diff --git a/portkey_ai/__init__.py b/portkey_ai/__init__.py
index e6ab0b31..1f538eb3 100644
--- a/portkey_ai/__init__.py
+++ b/portkey_ai/__init__.py
@@ -89,6 +89,10 @@
     AsyncVirtualKeys,
     Logs,
     AsyncLogs,
+    BetaRealtime,
+    AsyncBetaRealtime,
+    BetaSessions,
+    AsyncBetaSessions,
 )
 
 from portkey_ai.version import VERSION
@@ -197,4 +201,8 @@
     "AsyncVirtualKeys",
     "Logs",
     "AsyncLogs",
+    "BetaRealtime",
+    "AsyncBetaRealtime",
+    "BetaSessions",
+    "AsyncBetaSessions",
 ]
diff --git a/portkey_ai/_vendor/openai-1.40.1.dist-info/INSTALLER b/portkey_ai/_vendor/openai-1.58.1.dist-info/INSTALLER
similarity index 100%
rename from portkey_ai/_vendor/openai-1.40.1.dist-info/INSTALLER
rename to portkey_ai/_vendor/openai-1.58.1.dist-info/INSTALLER
diff --git a/portkey_ai/_vendor/openai-1.40.1.dist-info/METADATA b/portkey_ai/_vendor/openai-1.58.1.dist-info/METADATA
similarity index 77%
rename from portkey_ai/_vendor/openai-1.40.1.dist-info/METADATA
rename to portkey_ai/_vendor/openai-1.58.1.dist-info/METADATA
index be05c0d3..95f639ae 100644
--- a/portkey_ai/_vendor/openai-1.40.1.dist-info/METADATA
+++ b/portkey_ai/_vendor/openai-1.58.1.dist-info/METADATA
@@ -1,6 +1,6 @@
-Metadata-Version: 2.3
+Metadata-Version: 2.4
 Name: openai
-Version: 1.40.1
+Version: 1.58.1
 Summary: The official Python library for the openai API
 Project-URL: Homepage, https://github.com/openai/openai-python
 Project-URL: Repository, https://github.com/openai/openai-python
@@ -14,7 +14,6 @@ Classifier: Operating System :: Microsoft :: Windows
 Classifier: Operating System :: OS Independent
 Classifier: Operating System :: POSIX
 Classifier: Operating System :: POSIX :: Linux
-Classifier: Programming Language :: Python :: 3.7
 Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
@@ -22,9 +21,8 @@ Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Classifier: Typing :: Typed
-Requires-Python: >=3.7.1
+Requires-Python: >=3.8
 Requires-Dist: anyio<5,>=3.5.0
-Requires-Dist: cached-property; python_version < '3.8'
 Requires-Dist: distro<2,>=1.7.0
 Requires-Dist: httpx<1,>=0.23.0
 Requires-Dist: jiter<1,>=0.4.0
@@ -36,13 +34,15 @@ Provides-Extra: datalib
 Requires-Dist: numpy>=1; extra == 'datalib'
 Requires-Dist: pandas-stubs>=1.1.0.11; extra == 'datalib'
 Requires-Dist: pandas>=1.2.3; extra == 'datalib'
+Provides-Extra: realtime
+Requires-Dist: websockets<15,>=13; extra == 'realtime'
 Description-Content-Type: text/markdown
 
 # OpenAI Python API library
 
 [![PyPI version](https://img.shields.io/pypi/v/openai.svg)](https://pypi.org/project/openai/)
 
-The OpenAI Python library provides convenient access to the OpenAI REST API from any Python 3.7+
+The OpenAI Python library provides convenient access to the OpenAI REST API from any Python 3.8+
 application. The library includes type definitions for all request params and response fields,
 and offers both synchronous and asynchronous clients powered by [httpx](https://github.com/encode/httpx).
 
@@ -71,8 +71,7 @@ import os
 from openai import OpenAI
 
 client = OpenAI(
-    # This is the default and can be omitted
-    api_key=os.environ.get("OPENAI_API_KEY"),
+    api_key=os.environ.get("OPENAI_API_KEY"),  # This is the default and can be omitted
 )
 
 chat_completion = client.chat.completions.create(
@@ -82,7 +81,7 @@ chat_completion = client.chat.completions.create(
             "content": "Say this is a test",
         }
     ],
-    model="gpt-3.5-turbo",
+    model="gpt-4o",
 )
 ```
 
@@ -91,6 +90,48 @@ we recommend using [python-dotenv](https://pypi.org/project/python-dotenv/)
 to add `OPENAI_API_KEY="My API Key"` to your `.env` file
 so that your API Key is not stored in source control.
 
+### Vision
+
+With a hosted image:
+
+```python
+response = client.chat.completions.create(
+    model="gpt-4o-mini",
+    messages=[
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": prompt},
+                {
+                    "type": "image_url",
+                    "image_url": {"url": f"{img_url}"},
+                },
+            ],
+        }
+    ],
+)
+```
+
+With the image as a base64 encoded string:
+
+```python
+response = client.chat.completions.create(
+    model="gpt-4o-mini",
+    messages=[
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": prompt},
+                {
+                    "type": "image_url",
+                    "image_url": {"url": f"data:{img_type};base64,{img_b64_str}"},
+                },
+            ],
+        }
+    ],
+)
+```
+
 ### Polling Helpers
 
 When interacting with the API some actions such as starting a Run and adding files to vector stores are asynchronous and take time to complete. The SDK includes
@@ -151,8 +192,7 @@ import asyncio
 from openai import AsyncOpenAI
 
 client = AsyncOpenAI(
-    # This is the default and can be omitted
-    api_key=os.environ.get("OPENAI_API_KEY"),
+    api_key=os.environ.get("OPENAI_API_KEY"),  # This is the default and can be omitted
 )
 
 
@@ -164,7 +204,7 @@ async def main() -> None:
                 "content": "Say this is a test",
             }
         ],
-        model="gpt-3.5-turbo",
+        model="gpt-4o",
     )
 
 
@@ -183,8 +223,13 @@ from openai import OpenAI
 client = OpenAI()
 
 stream = client.chat.completions.create(
-    model="gpt-4",
-    messages=[{"role": "user", "content": "Say this is a test"}],
+    messages=[
+        {
+            "role": "user",
+            "content": "Say this is a test",
+        }
+    ],
+    model="gpt-4o",
     stream=True,
 )
 for chunk in stream:
@@ -194,6 +239,7 @@ for chunk in stream:
 The async client uses the exact same interface.
 
 ```python
+import asyncio
 from openai import AsyncOpenAI
 
 client = AsyncOpenAI()
@@ -230,7 +276,7 @@ openai.base_url = "https://..."
 openai.default_headers = {"x-foo": "true"}
 
 completion = openai.chat.completions.create(
-    model="gpt-4",
+    model="gpt-4o",
     messages=[
         {
             "role": "user",
@@ -252,6 +298,67 @@ We recommend that you always instantiate a client (e.g., with `client = OpenAI()
 - It's harder to mock for testing purposes
 - It's not possible to control cleanup of network connections
 
+## Realtime API beta
+
+The Realtime API enables you to build low-latency, multi-modal conversational experiences. It currently supports text and audio as both input and output, as well as [function calling](https://platform.openai.com/docs/guides/function-calling) through a WebSocket connection.
+
+Under the hood the SDK uses the [`websockets`](https://websockets.readthedocs.io/en/stable/) library to manage connections.
+
+The Realtime API works through a combination of client-sent events and server-sent events. Clients can send events to do things like update session configuration or send text and audio inputs. Server events confirm when audio responses have completed, or when a text response from the model has been received. A full event reference can be found [here](https://github.com/openai/openai-python/tree/main/platform.openai.com/docs/api-reference/realtime-client-events) and a guide can be found [here](https://platform.openai.com/docs/guides/realtime).
+
+Basic text based example:
+
+```py
+import asyncio
+from openai import AsyncOpenAI
+
+async def main():
+    client = AsyncOpenAI()
+
+    async with client.beta.realtime.connect(model="gpt-4o-realtime-preview-2024-10-01") as connection:
+        await connection.session.update(session={'modalities': ['text']})
+
+        await connection.conversation.item.create(
+            item={
+                "type": "message",
+                "role": "user",
+                "content": [{"type": "input_text", "text": "Say hello!"}],
+            }
+        )
+        await connection.response.create()
+
+        async for event in connection:
+            if event.type == 'response.text.delta':
+                print(event.delta, flush=True, end="")
+
+            elif event.type == 'response.text.done':
+                print()
+
+            elif event.type == "response.done":
+                break
+
+asyncio.run(main())
+```
+
+However the real magic of the Realtime API is handling audio inputs / outputs, see this example [TUI script](https://github.com/openai/openai-python/blob/main/examples/realtime/push_to_talk_app.py) for a fully fledged example.
+
+### Realtime error handling
+
+Whenever an error occurs, the Realtime API will send an [`error` event](https://platform.openai.com/docs/guides/realtime/realtime-api-beta#handling-errors) and the connection will stay open and remain usable. This means you need to handle it yourself, as *no errors are raised directly* by the SDK when an `error` event comes in.
+
+```py
+client = AsyncOpenAI()
+
+async with client.beta.realtime.connect(model="gpt-4o-realtime-preview-2024-10-01") as connection:
+    ...
+    async for event in connection:
+        if event.type == 'error':
+            print(event.error.type)
+            print(event.error.code)
+            print(event.error.event_id)
+            print(event.error.message)
+```
+
 ## Using types
 
 Nested request parameters are [TypedDicts](https://docs.python.org/3/library/typing.html#typing.TypedDict). Responses are [Pydantic models](https://docs.pydantic.dev) which also provide helper methods for things like:
@@ -348,7 +455,7 @@ completion = client.chat.completions.create(
             "content": "Can you generate an example json object describing a fruit?",
         }
     ],
-    model="gpt-3.5-turbo-1106",
+    model="gpt-4o",
     response_format={"type": "json_object"},
 )
 ```
@@ -388,7 +495,7 @@ client = OpenAI()
 
 try:
     client.fine_tuning.jobs.create(
-        model="gpt-3.5-turbo",
+        model="gpt-4o",
         training_file="file-abc123",
     )
 except openai.APIConnectionError as e:
@@ -415,6 +522,24 @@ Error codes are as followed:
 | >=500       | `InternalServerError`      |
 | N/A         | `APIConnectionError`       |
 
+## Request IDs
+
+> For more information on debugging requests, see [these docs](https://platform.openai.com/docs/api-reference/debugging-requests)
+
+All object responses in the SDK provide a `_request_id` property which is added from the `x-request-id` response header so that you can quickly log failing requests and report them back to OpenAI.
+
+```python
+completion = await client.chat.completions.create(
+    messages=[{"role": "user", "content": "Say this is a test"}], model="gpt-4"
+)
+print(completion._request_id)  # req_123
+```
+
+Note that unlike other properties that use an `_` prefix, the `_request_id` property
+*is* public. Unless documented otherwise, *all* other `_` prefix properties,
+methods and modules are *private*.
+
+
 ### Retries
 
 Certain errors are automatically retried 2 times by default, with a short exponential backoff.
@@ -437,10 +562,10 @@ client.with_options(max_retries=5).chat.completions.create(
     messages=[
         {
             "role": "user",
-            "content": "How can I get the name of the current day in Node.js?",
+            "content": "How can I get the name of the current day in JavaScript?",
         }
     ],
-    model="gpt-3.5-turbo",
+    model="gpt-4o",
 )
 ```
 
@@ -471,7 +596,7 @@ client.with_options(timeout=5.0).chat.completions.create(
             "content": "How can I list all files in a directory using Python?",
         }
     ],
-    model="gpt-3.5-turbo",
+    model="gpt-4o",
 )
 ```
 
@@ -485,12 +610,14 @@ Note that requests that time out are [retried twice by default](https://github.c
 
 We use the standard library [`logging`](https://docs.python.org/3/library/logging.html) module.
 
-You can enable logging by setting the environment variable `OPENAI_LOG` to `debug`.
+You can enable logging by setting the environment variable `OPENAI_LOG` to `info`.
 
 ```shell
-$ export OPENAI_LOG=debug
+$ export OPENAI_LOG=info
 ```
 
+Or to `debug` for more verbose logging.
+
 ### How to tell whether `None` means `null` or missing
 
 In an API response, a field may be explicitly `null`, or missing entirely; in either case, its value is `None` in this library. You can differentiate the two cases with `.model_fields_set`:
@@ -516,7 +643,7 @@ response = client.chat.completions.with_raw_response.create(
         "role": "user",
         "content": "Say this is a test",
     }],
-    model="gpt-3.5-turbo",
+    model="gpt-4o",
 )
 print(response.headers.get('X-My-Header'))
 
@@ -549,7 +676,7 @@ with client.chat.completions.with_streaming_response.create(
             "content": "Say this is a test",
         }
     ],
-    model="gpt-3.5-turbo",
+    model="gpt-4o",
 ) as response:
     print(response.headers.get("X-My-Header"))
 
@@ -598,18 +725,19 @@ can also get all the extra fields on the Pydantic model as a dict with
 
 You can directly override the [httpx client](https://www.python-httpx.org/api/#client) to customize it for your use case, including:
 
-- Support for proxies
-- Custom transports
+- Support for [proxies](https://www.python-httpx.org/advanced/proxies/)
+- Custom [transports](https://www.python-httpx.org/advanced/transports/)
 - Additional [advanced](https://www.python-httpx.org/advanced/clients/) functionality
 
 ```python
+import httpx
 from openai import OpenAI, DefaultHttpxClient
 
 client = OpenAI(
     # Or use the `OPENAI_BASE_URL` env var
-    base_url="http://my.test.server.example.com:8083",
+    base_url="http://my.test.server.example.com:8083/v1",
     http_client=DefaultHttpxClient(
-        proxies="http://my.test.proxy.example.com",
+        proxy="http://my.test.proxy.example.com",
         transport=httpx.HTTPTransport(local_address="0.0.0.0"),
     ),
 )
@@ -625,6 +753,16 @@ client.with_options(http_client=DefaultHttpxClient(...))
 
 By default the library closes underlying HTTP connections whenever the client is [garbage collected](https://docs.python.org/3/reference/datamodel.html#object.__del__). You can manually close the client using the `.close()` method if desired, or with a context manager that closes when exiting.
 
+```py
+from openai import OpenAI
+
+with OpenAI() as client:
+  # make requests here
+  ...
+
+# HTTP client is now closed
+```
+
 ## Microsoft Azure OpenAI
 
 To use this library with [Azure OpenAI](https://learn.microsoft.com/azure/ai-services/openai/overview), use the `AzureOpenAI`
@@ -679,6 +817,21 @@ We take backwards-compatibility seriously and work hard to ensure you can rely o
 
 We are keen for your feedback; please open an [issue](https://www.github.com/openai/openai-python/issues) with questions, bugs, or suggestions.
 
+### Determining the installed version
+
+If you've upgraded to the latest version but aren't seeing any new features you were expecting then your python environment is likely still using an older version.
+
+You can determine the version that is being used at runtime with:
+
+```py
+import openai
+print(openai.__version__)
+```
+
 ## Requirements
 
-Python 3.7 or higher.
+Python 3.8 or higher.
+
+## Contributing
+
+See [the contributing documentation](https://github.com/openai/openai-python/tree/main/./CONTRIBUTING.md).
diff --git a/portkey_ai/_vendor/openai-1.40.1.dist-info/RECORD b/portkey_ai/_vendor/openai-1.58.1.dist-info/RECORD
similarity index 50%
rename from portkey_ai/_vendor/openai-1.40.1.dist-info/RECORD
rename to portkey_ai/_vendor/openai-1.58.1.dist-info/RECORD
index 573784d2..1ef0ad0d 100644
--- a/portkey_ai/_vendor/openai-1.40.1.dist-info/RECORD
+++ b/portkey_ai/_vendor/openai-1.58.1.dist-info/RECORD
@@ -1,347 +1,440 @@
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/__main__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_base_client.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_client.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_compat.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_constants.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_exceptions.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_extras/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_extras/_common.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_extras/numpy_proxy.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_extras/pandas_proxy.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_files.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_legacy_response.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_models.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_module_client.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_qs.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_resource.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_response.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_streaming.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_types.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_utils/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_utils/_logs.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_utils/_proxy.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_utils/_reflection.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_utils/_streams.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_utils/_sync.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_utils/_transform.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_utils/_typing.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_utils/_utils.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/_version.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_api/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_api/_main.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_api/audio.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_api/chat/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_api/chat/completions.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_api/completions.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_api/files.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_api/image.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_api/models.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_cli.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_errors.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_models.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_progress.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_tools/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_tools/_main.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_tools/fine_tunes.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_tools/migrate.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/cli/_utils.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/_old_api.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/_parsing/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/_parsing/_completions.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/_pydantic.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/_tools.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/_validators.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/azure.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/streaming/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/streaming/_assistants.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/streaming/_deltas.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/streaming/chat/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/streaming/chat/_completions.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/streaming/chat/_events.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/lib/streaming/chat/_types.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/pagination.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/audio/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/audio/audio.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/audio/speech.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/audio/transcriptions.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/audio/translations.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/batches.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/assistants.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/beta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/chat/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/chat/chat.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/chat/completions.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/threads/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/threads/messages.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/threads/runs/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/threads/runs/runs.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/threads/runs/steps.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/threads/threads.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/vector_stores/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/vector_stores/file_batches.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/vector_stores/files.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/beta/vector_stores/vector_stores.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/chat/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/chat/chat.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/chat/completions.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/completions.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/embeddings.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/files.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/fine_tuning/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/fine_tuning/fine_tuning.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/fine_tuning/jobs/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/fine_tuning/jobs/checkpoints.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/fine_tuning/jobs/jobs.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/images.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/models.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/moderations.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/uploads/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/uploads/parts.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/resources/uploads/uploads.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/audio/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/audio/speech_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/audio/speech_model.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/audio/transcription.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/audio/transcription_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/audio/translation.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/audio/translation_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/audio_model.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/batch.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/batch_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/batch_error.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/batch_list_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/batch_request_counts.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_deleted.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_list_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_response_format_option.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_response_format_option_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_stream_event.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_tool.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_tool_choice.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_tool_choice_function.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_tool_choice_function_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_tool_choice_option.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_tool_choice_option_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_tool_choice_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_tool_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/assistant_update_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/chat/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/code_interpreter_tool.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/code_interpreter_tool_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/file_search_tool.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/file_search_tool_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/function_tool.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/function_tool_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/thread.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/thread_create_and_run_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/thread_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/thread_deleted.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/thread_update_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/annotation.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/annotation_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/file_citation_annotation.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/file_citation_delta_annotation.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/file_path_annotation.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/file_path_delta_annotation.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_file.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_file_content_block.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_file_content_block_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_file_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_file_delta_block.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_file_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_url.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_url_content_block.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_url_content_block_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_url_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_url_delta_block.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/image_url_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/message.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/message_content.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/message_content_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/message_content_part_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/message_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/message_deleted.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/message_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/message_delta_event.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/message_list_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/message_update_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/refusal_content_block.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/refusal_delta_block.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/required_action_function_tool_call.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/run.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/run_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/run_list_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/run_status.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/run_submit_tool_outputs_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/run_update_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/code_interpreter_logs.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/code_interpreter_output_image.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/code_interpreter_tool_call.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/code_interpreter_tool_call_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/file_search_tool_call.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/file_search_tool_call_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/function_tool_call.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/function_tool_call_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/message_creation_step_details.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/run_step.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/run_step_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/run_step_delta_event.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/run_step_delta_message_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/step_list_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/tool_call.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/tool_call_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/tool_call_delta_object.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/runs/tool_calls_step_details.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/text.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/text_content_block.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/text_content_block_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/text_delta.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/threads/text_delta_block.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_store.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_store_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_store_deleted.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_store_list_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_store_update_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_stores/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_stores/file_batch_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_stores/file_batch_list_files_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_stores/file_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_stores/file_list_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_stores/vector_store_file.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_stores/vector_store_file_batch.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/beta/vector_stores/vector_store_file_deleted.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_assistant_message_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_chunk.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_content_part_image_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_content_part_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_content_part_refusal_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_content_part_text_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_function_call_option_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_function_message_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_message.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_message_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_message_tool_call.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_message_tool_call_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_named_tool_choice_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_role.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_stream_options_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_system_message_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_token_logprob.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_tool_choice_option_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_tool_message_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_tool_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/chat_completion_user_message_param.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/completion_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/parsed_chat_completion.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat/parsed_function_tool_call.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/chat_model.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/completion.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/completion_choice.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/completion_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/completion_usage.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/create_embedding_response.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/embedding.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/embedding_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/file_content.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/file_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/file_deleted.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/file_list_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/file_object.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/fine_tuning_job.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/fine_tuning_job_event.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/fine_tuning_job_integration.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/fine_tuning_job_wandb_integration.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/fine_tuning_job_wandb_integration_object.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/job_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/job_list_events_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/job_list_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/jobs/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/jobs/checkpoint_list_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/fine_tuning/jobs/fine_tuning_job_checkpoint.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/image.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/image_create_variation_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/image_edit_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/image_generate_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/image_model.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/images_response.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/model.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/model_deleted.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/moderation.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/moderation_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/moderation_create_response.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/moderation_model.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared/error_object.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared/function_definition.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared/function_parameters.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared/response_format_json_object.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared/response_format_json_schema.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared/response_format_text.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared_params/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared_params/function_definition.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared_params/function_parameters.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared_params/response_format_json_object.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared_params/response_format_json_schema.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/shared_params/response_format_text.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/upload.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/upload_complete_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/upload_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/uploads/__init__.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/uploads/part_create_params.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/types/uploads/upload_part.cpython-39.pyc,,
-../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-plhz866u/lib/python/openai/version.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/__main__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_base_client.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_client.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_compat.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_constants.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_exceptions.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_extras/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_extras/_common.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_extras/numpy_proxy.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_extras/pandas_proxy.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_files.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_legacy_response.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_models.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_module_client.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_qs.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_resource.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_response.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_streaming.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_types.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_utils/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_utils/_logs.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_utils/_proxy.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_utils/_reflection.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_utils/_streams.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_utils/_sync.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_utils/_transform.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_utils/_typing.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_utils/_utils.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/_version.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_api/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_api/_main.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_api/audio.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_api/chat/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_api/chat/completions.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_api/completions.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_api/files.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_api/image.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_api/models.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_cli.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_errors.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_models.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_progress.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_tools/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_tools/_main.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_tools/fine_tunes.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_tools/migrate.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/cli/_utils.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/_old_api.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/_parsing/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/_parsing/_completions.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/_pydantic.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/_tools.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/_validators.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/azure.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/streaming/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/streaming/_assistants.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/streaming/_deltas.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/streaming/chat/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/streaming/chat/_completions.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/streaming/chat/_events.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/lib/streaming/chat/_types.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/pagination.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/audio/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/audio/audio.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/audio/speech.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/audio/transcriptions.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/audio/translations.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/batches.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/assistants.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/beta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/chat/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/chat/chat.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/chat/completions.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/realtime/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/realtime/realtime.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/realtime/sessions.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/threads/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/threads/messages.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/threads/runs/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/threads/runs/runs.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/threads/runs/steps.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/threads/threads.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/vector_stores/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/vector_stores/file_batches.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/vector_stores/files.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/beta/vector_stores/vector_stores.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/chat/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/chat/chat.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/chat/completions.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/completions.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/embeddings.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/files.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/fine_tuning/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/fine_tuning/fine_tuning.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/fine_tuning/jobs/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/fine_tuning/jobs/checkpoints.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/fine_tuning/jobs/jobs.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/images.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/models.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/moderations.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/uploads/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/uploads/parts.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/resources/uploads/uploads.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/speech_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/speech_model.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/transcription.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/transcription_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/transcription_create_response.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/transcription_segment.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/transcription_verbose.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/transcription_word.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/translation.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/translation_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/translation_create_response.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio/translation_verbose.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio_model.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/audio_response_format.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/batch.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/batch_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/batch_error.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/batch_list_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/batch_request_counts.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_deleted.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_list_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_response_format_option.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_response_format_option_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_stream_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_tool.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_tool_choice.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_tool_choice_function.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_tool_choice_function_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_tool_choice_option.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_tool_choice_option_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_tool_choice_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_tool_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/assistant_update_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/auto_file_chunking_strategy_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/chat/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/code_interpreter_tool.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/code_interpreter_tool_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/file_chunking_strategy.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/file_chunking_strategy_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/file_search_tool.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/file_search_tool_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/function_tool.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/function_tool_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/other_file_chunking_strategy_object.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_created_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_content.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_content_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_create_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_create_event_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_created_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_delete_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_delete_event_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_deleted_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_input_audio_transcription_completed_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_input_audio_transcription_failed_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_truncate_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_truncate_event_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/conversation_item_truncated_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/error_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/input_audio_buffer_append_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/input_audio_buffer_append_event_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/input_audio_buffer_clear_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/input_audio_buffer_clear_event_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/input_audio_buffer_cleared_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/input_audio_buffer_commit_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/input_audio_buffer_commit_event_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/input_audio_buffer_committed_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/input_audio_buffer_speech_started_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/input_audio_buffer_speech_stopped_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/rate_limits_updated_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/realtime_client_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/realtime_client_event_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/realtime_connect_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/realtime_response.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/realtime_response_status.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/realtime_response_usage.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/realtime_server_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_audio_delta_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_audio_done_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_audio_transcript_delta_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_audio_transcript_done_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_cancel_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_cancel_event_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_content_part_added_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_content_part_done_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_create_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_create_event_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_created_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_done_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_function_call_arguments_delta_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_function_call_arguments_done_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_output_item_added_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_output_item_done_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_text_delta_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/response_text_done_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/session.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/session_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/session_create_response.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/session_created_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/session_update_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/session_update_event_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/realtime/session_updated_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/static_file_chunking_strategy.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/static_file_chunking_strategy_object.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/static_file_chunking_strategy_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/thread.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/thread_create_and_run_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/thread_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/thread_deleted.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/thread_update_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/annotation.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/annotation_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/file_citation_annotation.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/file_citation_delta_annotation.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/file_path_annotation.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/file_path_delta_annotation.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_file.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_file_content_block.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_file_content_block_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_file_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_file_delta_block.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_file_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_url.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_url_content_block.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_url_content_block_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_url_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_url_delta_block.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/image_url_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/message.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/message_content.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/message_content_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/message_content_part_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/message_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/message_deleted.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/message_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/message_delta_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/message_list_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/message_update_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/refusal_content_block.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/refusal_delta_block.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/required_action_function_tool_call.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/run.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/run_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/run_list_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/run_status.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/run_submit_tool_outputs_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/run_update_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/code_interpreter_logs.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/code_interpreter_output_image.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/code_interpreter_tool_call.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/code_interpreter_tool_call_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/file_search_tool_call.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/file_search_tool_call_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/function_tool_call.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/function_tool_call_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/message_creation_step_details.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/run_step.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/run_step_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/run_step_delta_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/run_step_delta_message_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/run_step_include.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/step_list_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/step_retrieve_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/tool_call.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/tool_call_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/tool_call_delta_object.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/runs/tool_calls_step_details.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/text.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/text_content_block.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/text_content_block_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/text_delta.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/threads/text_delta_block.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_store.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_store_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_store_deleted.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_store_list_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_store_update_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_stores/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_stores/file_batch_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_stores/file_batch_list_files_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_stores/file_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_stores/file_list_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_stores/vector_store_file.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_stores/vector_store_file_batch.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/beta/vector_stores/vector_store_file_deleted.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_assistant_message_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_audio.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_audio_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_chunk.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_content_part_image_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_content_part_input_audio_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_content_part_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_content_part_refusal_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_content_part_text_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_developer_message_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_function_call_option_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_function_message_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_message.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_message_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_message_tool_call.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_message_tool_call_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_modality.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_named_tool_choice_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_prediction_content_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_reasoning_effort.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_role.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_stream_options_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_system_message_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_token_logprob.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_tool_choice_option_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_tool_message_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_tool_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/chat_completion_user_message_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/completion_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/parsed_chat_completion.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat/parsed_function_tool_call.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/chat_model.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/completion.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/completion_choice.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/completion_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/completion_usage.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/create_embedding_response.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/embedding.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/embedding_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/embedding_model.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/file_content.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/file_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/file_deleted.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/file_list_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/file_object.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/file_purpose.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/fine_tuning_job.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/fine_tuning_job_event.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/fine_tuning_job_integration.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/fine_tuning_job_wandb_integration.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/fine_tuning_job_wandb_integration_object.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/job_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/job_list_events_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/job_list_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/jobs/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/jobs/checkpoint_list_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/fine_tuning/jobs/fine_tuning_job_checkpoint.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/image.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/image_create_variation_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/image_edit_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/image_generate_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/image_model.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/images_response.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/model.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/model_deleted.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/moderation.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/moderation_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/moderation_create_response.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/moderation_image_url_input_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/moderation_model.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/moderation_multi_modal_input_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/moderation_text_input_param.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared/error_object.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared/function_definition.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared/function_parameters.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared/response_format_json_object.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared/response_format_json_schema.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared/response_format_text.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared_params/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared_params/function_definition.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared_params/function_parameters.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared_params/response_format_json_object.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared_params/response_format_json_schema.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/shared_params/response_format_text.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/upload.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/upload_complete_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/upload_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/uploads/__init__.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/uploads/part_create_params.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/uploads/upload_part.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/types/websocket_connection_options.cpython-39.pyc,,
+../../../../../../../../../Users/chandeep/Library/Caches/com.apple.python/private/var/folders/9_/nx5thgw547s1zlwxzhbjtb1m0000gn/T/pip-target-_6i4wnmx/lib/python/openai/version.cpython-39.pyc,,
 ../../bin/openai,sha256=hl2E5BbKWVfkczcW65l8G1zyeJ3Si5m9TUnp5aG8gtY,276
-openai-1.40.1.dist-info/INSTALLER,sha256=zuuue4knoyJ-UwPPXg8fezS7VCrXJQrAP7zeNuwvFQg,4
-openai-1.40.1.dist-info/METADATA,sha256=E4y7Ouv_ZSZBT7WuW6u1fIcdVyrMVjYVgunbAtMATcw,22163
-openai-1.40.1.dist-info/RECORD,,
-openai-1.40.1.dist-info/REQUESTED,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-openai-1.40.1.dist-info/WHEEL,sha256=1yFddiXMmvYK7QYTqtRNtX66WJ0Mz8PYEiEUoOUUxRY,87
-openai-1.40.1.dist-info/entry_points.txt,sha256=kAYhQEmziJwsKs5raYAIOvJ2LWmbz5dulEXOzsY71ro,43
-openai-1.40.1.dist-info/licenses/LICENSE,sha256=d0M6HDjQ76tf255XPlAGkIoECMe688MXcGEYsOFySfI,11336
-openai/__init__.py,sha256=YhCuMuxZHoRn6BnOxawEFt8fRZPnhBWGongW3CP-F3k,10191
+openai-1.58.1.dist-info/INSTALLER,sha256=zuuue4knoyJ-UwPPXg8fezS7VCrXJQrAP7zeNuwvFQg,4
+openai-1.58.1.dist-info/METADATA,sha256=bMFx74-7aYzZDEuWPpWEycqxo8YvoA5fXXk4sCy6ygU,27285
+openai-1.58.1.dist-info/RECORD,,
+openai-1.58.1.dist-info/REQUESTED,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+openai-1.58.1.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+openai-1.58.1.dist-info/entry_points.txt,sha256=kAYhQEmziJwsKs5raYAIOvJ2LWmbz5dulEXOzsY71ro,43
+openai-1.58.1.dist-info/licenses/LICENSE,sha256=d0M6HDjQ76tf255XPlAGkIoECMe688MXcGEYsOFySfI,11336
+openai/__init__.py,sha256=HAeDtlHRw1QQs5BIDoE0k7bA7xOP5m7F4tYjXt_9c64,10209
 openai/__main__.py,sha256=bYt9eEaoRQWdejEHFD8REx9jxVEdZptECFsV7F49Ink,30
-openai/_base_client.py,sha256=_ywUgO9OA-ORypQlYrUUgHgUgGFYPMEWHW0DLLCVUBc,67506
-openai/_client.py,sha256=PXHky30KYjUMIH8WV7PjKcOAULO9-36AbN8y1DCFu70,22233
-openai/_compat.py,sha256=DvCJ4A6g0azYhKPpLKXoi1eYXpwhukMGUxJclk2DtUA,6832
+openai/_base_client.py,sha256=uWXGJEkKDgXxPEpz0-G5yjuJR28FTIhBEh6ijOiEWMA,69144
+openai/_client.py,sha256=FJRGkrdpHAFV2TOs04tO5uyKCA-cudlk4BlvCX3KI3Q,23355
+openai/_compat.py,sha256=Mtzi28qOK99ZBPcGcQqdjoUFk2MzzpqjaafjuwQ4NO0,6982
 openai/_constants.py,sha256=L1pfEhuz_wM2w2_U9P_9JZzTbrN4pbLo207l96rtKcQ,469
-openai/_exceptions.py,sha256=QoeAOAeOatAOxt80wJShYUQti-QQc2Pur45IxiWN-zk,4376
+openai/_exceptions.py,sha256=2BEuXwqce9z7X6lWLLXRqg1vOay_q-OdLz9lcj6Pluw,4798
 openai/_extras/__init__.py,sha256=LZbJLZ7aFHRcI7uiY4-wFQTdMp-BF6FER1QMhKVFkWk,107
 openai/_extras/_common.py,sha256=NWWtgbdJsO3hQGQxaXGfVk0LjeIE5AFZ8VS_795hhMc,364
 openai/_extras/numpy_proxy.py,sha256=hwZXa_JBAPD5taRhor1tGxK26g5IaK52JclQDl-dky0,799
 openai/_extras/pandas_proxy.py,sha256=NCEt1Dqwc_0H85YdsWPDE3lPDJtYnBT8G-gJE_BCeEc,637
 openai/_files.py,sha256=WEf6hxJN1u3pVkdnPCpinhxCUnOV2olt4J6vLoJ_k48,3616
-openai/_legacy_response.py,sha256=7NyYX04zAjuGGypOjOWwS7hljQq5OSyR-6B6_w3_q6g,15781
-openai/_models.py,sha256=3METTz2oSlPfVcHL0fZ-L5IPnvckZDkgEe1H58mRsFE,28210
+openai/_legacy_response.py,sha256=YBL2OTX7W139lVpcVHnNTsHRPNJxWHBAw6ZZHqnL2fs,16046
+openai/_models.py,sha256=k3sTshX4q5XbZy__I4bchWrYAtID1alxVdGOn_4jGQ4,30343
 openai/_module_client.py,sha256=gF_2bbdosIwUt29sQgrQRJOgNREvXF-IDxe4XKGhHjY,2523
 openai/_qs.py,sha256=AOkSz4rHtK4YI3ZU_kzea-zpwBUgEY8WniGmTPyEimc,4846
 openai/_resource.py,sha256=IQihFzFLhGOiGSlT2dO1ESWSTg2XypgbtAldtGdTOqU,1100
-openai/_response.py,sha256=UW8TM-E4YE6UzhKcWOoGvBH3cVKh-aFe1yEL7wZaMIQ,29018
+openai/_response.py,sha256=Juwnj0AMWnHc8HDjtdcQQpMIDyX170hzZPXaAK1e9Qw,29387
 openai/_streaming.py,sha256=t1UZrg53fVJB5Rs6k2sT9PBbvjp-IGrQzUq_5nlxKG4,13102
-openai/_types.py,sha256=77A36sAUMgrgTX3zNo2NKU_wbQZgoZWjGTwf3GTOGTc,6202
-openai/_utils/__init__.py,sha256=Uzq1-FIih_VUjzdNVWXks0sdC39KBKLMrZoz-_JOjJ4,1988
-openai/_utils/_logs.py,sha256=sFA_NejuNObTGGbfsXC03I38mrT9HjsgAJx4d3GP0ok,774
+openai/_types.py,sha256=GxKqy9_2_AUqbaRROzqhCJ47a7c-q_T6Bu8kV9a2qhA,6242
+openai/_utils/__init__.py,sha256=0yN65NkadWbqEFhfcHlJje-9jEHBC14ibHb9nC4ClOg,2130
+openai/_utils/_logs.py,sha256=IC5iwPflwelNpJEpWsvK3up-pol5hR8k_VL9fSukk_Y,1351
 openai/_utils/_proxy.py,sha256=z3zsateHtb0EARTWKk8QZNHfPkqJbqwd1lM993LBwGE,1902
-openai/_utils/_reflection.py,sha256=ZmGkIgT_PuwedyNBrrKGbxoWtkpytJNU1uU4QHnmEMU,1364
+openai/_utils/_reflection.py,sha256=aTXm-W0Kww4PJo5LPkUnQ92N-2UvrK1-D67cJVBlIgw,1426
 openai/_utils/_streams.py,sha256=SMC90diFFecpEg_zgDRVbdR3hSEIgVVij4taD-noMLM,289
-openai/_utils/_sync.py,sha256=9ex9pfOyd8xAF1LxpFx4IkqL8k0vk8srE2Ee-OTMQ0A,2840
-openai/_utils/_transform.py,sha256=NCz3q9_O-vuj60xVe-qzhEQ8uJWlZWJTsM-GwHDccf8,12958
-openai/_utils/_typing.py,sha256=tFbktdpdHCQliwzGsWysgn0P5H0JRdagkZdb_LegGkY,3838
-openai/_utils/_utils.py,sha256=LMVTMZG8pfu8AkJNSfmv_z3guQlOfm2UxDTjTTXggfg,11411
-openai/_version.py,sha256=byQHxbhaxiwEt7VT71Cengl03CfF1u54tElKLb-aRqM,159
+openai/_utils/_sync.py,sha256=03JeD-UR_e2O8dJEtD-v4zcyhlEpFkrcH8bgrSJMrxI,2437
+openai/_utils/_transform.py,sha256=Dkkyr7OveGmOolepcvXmVJWE3kqim4b0nM0h7yWbgeY,13468
+openai/_utils/_typing.py,sha256=nTJz0jcrQbEgxwy4TtAkNxuU0QHHlmc6mQtA6vIR8tg,4501
+openai/_utils/_utils.py,sha256=8UmbPOy_AAr2uUjjFui-VZSrVBHRj6bfNEKRp5YZP2A,12004
+openai/_version.py,sha256=DKLfLGiThHOxb7_mxxvh3kOxCAzNPvebOIc6-n1St50,159
 openai/cli/__init__.py,sha256=soGgtqyomgddl92H0KJRqHqGuaXIaghq86qkzLuVp7U,31
 openai/cli/_api/__init__.py,sha256=cj92MZq-9_1PQM8A4TQVsqKn5mcTDAGxHllJ0UvJOPE,58
 openai/cli/_api/_main.py,sha256=5yyfLURqCEaAN8B61gHaqVAaYgtyb9Xq0ncQ3P2BAh0,451
@@ -352,95 +445,105 @@ openai/cli/_api/completions.py,sha256=ysOmnbXpFz3VB5N_5USPdObiYew62vEn6rMtNFwTJG
 openai/cli/_api/files.py,sha256=6nKXFnsC2QE0bGnVUAG7BTLSu6K1_MhPE0ZJACmzgRY,2345
 openai/cli/_api/image.py,sha256=ovBExdn8oUK9ImOpsPafesfAlmcftLP2p7d37hcUtKU,5062
 openai/cli/_api/models.py,sha256=pGmIGZToj3raGGpKvPSq_EVUR-dqg4Vi0PNfZH98D2E,1295
-openai/cli/_cli.py,sha256=WxqTnhVVtfzX0z7hV5fcvd3hkihaUgwOWpXOwyCS4Fc,6743
+openai/cli/_cli.py,sha256=o6zWCnq84u-DIGZuR9YoOUxTGTpx-oCU5mgAKDi555c,6779
 openai/cli/_errors.py,sha256=nejlu1HnOyAIr2n7uqpFtWn8XclWj_9N8FwgfT3BPK8,471
 openai/cli/_models.py,sha256=tgsldjG216KpwgAZ5pS0sV02FQvONDJU2ElA4kCCiIU,491
 openai/cli/_progress.py,sha256=aMLssU9jh-LoqRYH3608jNos7r6vZKnHTRlHxFznzv4,1406
 openai/cli/_tools/__init__.py,sha256=cj92MZq-9_1PQM8A4TQVsqKn5mcTDAGxHllJ0UvJOPE,58
 openai/cli/_tools/_main.py,sha256=pakjEXHRHqYlTml-RxV7fNrRtRXzmZBinoPi1AJipFY,467
 openai/cli/_tools/fine_tunes.py,sha256=RQgYMzifk6S7Y1I1K6huqco2QxmXa7gVUlHl6SrKTSU,1543
-openai/cli/_tools/migrate.py,sha256=GD3zHR700FRIhdx3gBqIrRLPrKjx4pDAKUgvnO0J2ug,5013
+openai/cli/_tools/migrate.py,sha256=o-iomzhtC6N6X5H5GDlgQ_QOaIovE2YA9oHc_tIAUj8,4497
 openai/cli/_utils.py,sha256=oiTc9MnxQh_zxAZ1OIHPkoDpCll0NF9ZgkdFHz4T-Bs,848
 openai/lib/.keep,sha256=wuNrz-5SXo3jJaJOJgz4vFHM41YH_g20F5cRQo0vLes,224
 openai/lib/__init__.py,sha256=BMTfMnlbugMgDA1STDIAlx4bI4t4l_8bQmJxd0th0n8,126
 openai/lib/_old_api.py,sha256=XZnXBrEKuTd70iJirj5mGW35fZoqruJobbBTq6bvg10,1947
 openai/lib/_parsing/__init__.py,sha256=wS3BYvMGj9TqiPqOe3rO1sleaAJqHVuCaQuCE5rZIUw,539
-openai/lib/_parsing/_completions.py,sha256=fXPhBD67s50THskYzBeueqo7B77toGRfZ3oaYv_tKVE,8601
-openai/lib/_pydantic.py,sha256=_bT7JPIsCTuznyA4GnyRE6tbmw9eIfjQ6wBsVAR7Xn4,2364
+openai/lib/_parsing/_completions.py,sha256=I1KpjdI9p8Me-nsLF2szjEYF_7x4k28WGH5GdZeKpzI,9138
+openai/lib/_pydantic.py,sha256=ndHdDDSEGg8Jbhc7JvLQHiIrZwLR36bCcUAlzwLmOdk,5282
 openai/lib/_tools.py,sha256=xrzM7jNgehZGsRQ9kSgn1q33z9cHrgf0b8UMo5wrTFw,1501
 openai/lib/_validators.py,sha256=cXJXFuaAl7jeJcYHXXnFa4NHGtHs-_zt3Zs1VVCmQo4,35288
-openai/lib/azure.py,sha256=iaiopzv8xI_JlYdtgEmnu32krLm0YiG44cDs_ictlgw,21536
+openai/lib/azure.py,sha256=s2jrnQmfvlpNREe4A7QcmeXeIuuI9zIWG3ltyfvuQOM,22450
 openai/lib/streaming/__init__.py,sha256=kD3LpjsqU7caDQDhB-YjTUl9qqbb5sPnGGSI2yQYC70,379
-openai/lib/streaming/_assistants.py,sha256=_t1R-cTCXmKfQ3aLwSqGnYqjOxJIWhoWSvDHqIFCaPw,40575
+openai/lib/streaming/_assistants.py,sha256=LUWSinmYopQIkQ5xSg73b6BWbkRkQS5JvX62w_V9xSw,40692
 openai/lib/streaming/_deltas.py,sha256=I7B_AznXZwlBmE8Puau7ayTQUx6hMIEVE8FYTQm2fjs,2502
-openai/lib/streaming/chat/__init__.py,sha256=d243EsKxxHQ_MpUxecmYdLy4ZRVY6BKhL6QNSfLdtRY,1245
-openai/lib/streaming/chat/_completions.py,sha256=Kje6_zfjFKxk0NIOSc9JZ_YxkhLKSlb8-weEtzt93Sc,28604
+openai/lib/streaming/chat/__init__.py,sha256=7krL_atOvvpQkY_byWSglSfDsMs5hdoxHmz4Ulq7lcc,1305
+openai/lib/streaming/chat/_completions.py,sha256=icXzr6TwaQvOOEZHRLIfw106YVUT9mLGjQt6QJ1ObKI,29944
 openai/lib/streaming/chat/_events.py,sha256=lstVmM6YR2Cs9drikzrY9JCZn9Nbfym0aKIPtNpxL6w,2618
 openai/lib/streaming/chat/_types.py,sha256=-SYVBNhGkOUoJ-8dotxpCRqPJpfyOQ8hwR2_HrsQCRI,739
 openai/pagination.py,sha256=B9ejXEAR_hYGLHfqb9xEEsE0u5dCUMjvplOce5dpY7M,2760
 openai/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 openai/resources/__init__.py,sha256=eYonVyf6AAmk-b8JYSYmo5EEMv89ovxiAY5A83ti8J8,4533
 openai/resources/audio/__init__.py,sha256=YM7FHvPKVlj_v6EIgfpUQsb6q4hS2hVQ3gfkgic0sP0,1687
-openai/resources/audio/audio.py,sha256=1HHcDRWT58KshYelRdSnJs-0bvMBRS1vOhnU-h_oP5s,4481
-openai/resources/audio/speech.py,sha256=s93RA77J_uS1Nu95dnrbbIahE-U-Csr9RozgqMFxB54,7885
-openai/resources/audio/transcriptions.py,sha256=acLwtVFCuTuG4OYHdzJlRq70hAMLzWV8YJGGc64EBOU,11075
-openai/resources/audio/translations.py,sha256=QFrFjB3JkruF7MPCJfsgGDRSlyKHAEdJLaF-OdhR3UY,8979
-openai/resources/batches.py,sha256=XuXe_9xb_x-Kq0myKTkcKHQqCf3OHhhNIXr4295mNHM,18560
+openai/resources/audio/audio.py,sha256=MMJHbfXmyYmQU7dF8XsD0YOIqdlG3gtxUqTihOuVx8o,5499
+openai/resources/audio/speech.py,sha256=yPoi_Xozv0Yuikbf2dxhAyRdN2q_sWDQoHNCxUayC-E,8903
+openai/resources/audio/transcriptions.py,sha256=4X71pe1lvelNRPSlHy2jAtIMyETYwWieLShBdr12MN0,18507
+openai/resources/audio/translations.py,sha256=4Y-ognKnSi72qhwX8FCKB-5JhvaAS2Wnq2ivTFmpUoU,15711
+openai/resources/batches.py,sha256=8wb-oy81IkxpABjT_11JKP7nzTmGmP35lD6WGecWmn8,19578
 openai/resources/beta/__init__.py,sha256=nXoV4P8WCrbEZuNMtptbIuy_LqlVafY9lJ2qfW35GFc,1636
-openai/resources/beta/assistants.py,sha256=2zxpTMThsrleTO_jQiHyvJT0mu2Tth2pS5DKpO5B2UI,39567
-openai/resources/beta/beta.py,sha256=Ys5tPrAmvr7_CmHJLKhnsQqWXGo5seX8-rP8HhXwk0w,4702
+openai/resources/beta/assistants.py,sha256=j1BE3q4aCGzridJ8wyhzn0FeI3Gvy56jRK57EA-SuXk,40533
+openai/resources/beta/beta.py,sha256=D9mhIg_Qc0tUq23AVRUI6Z1WRF_ekeNG5sHeRYyhFXk,6602
 openai/resources/beta/chat/__init__.py,sha256=d_fpyFMAG3iRAPIXANPfRG4HtEm6U_uMUYep7Skj2uY,263
 openai/resources/beta/chat/chat.py,sha256=sNvU8Fi_o3dWkD_X4Mobafv9XWBP6Y2dJxng-NdFXUs,597
-openai/resources/beta/chat/completions.py,sha256=L1G6JsVjXT1NoWJdE6Z_xdjc1MXh69gppipaHG4LO94,20280
+openai/resources/beta/chat/completions.py,sha256=Z_x_hxpemrmROMrfyx6dUALppPuqNgswgW9YQ3ngHYI,28553
+openai/resources/beta/realtime/__init__.py,sha256=0TBjHlLRsG-hudbiE8f-EXETNkDRAxqkCVAgODiUnYo,862
+openai/resources/beta/realtime/realtime.py,sha256=mbzDEKD90pt06pBg_A44t-xZnQTInwP83WuOlwgBDTY,36992
+openai/resources/beta/realtime/sessions.py,sha256=i53-QVMaqK3sGP22gh250kANFlRbP4V-g8uffWzKHS8,16093
 openai/resources/beta/threads/__init__.py,sha256=fQ_qdUVSfouVS5h47DlTb5mamChT4K-v-siPuuAB6do,1177
-openai/resources/beta/threads/messages.py,sha256=hD_f8FOJ-qfkByJzeZv-YxaNvxUIVKfsEf1HNy_IEOo,26346
+openai/resources/beta/threads/messages.py,sha256=LBjgJAK-0g_lkhIX2WG6qNT0RzSTknO0nRlqkVQw-B8,27372
 openai/resources/beta/threads/runs/__init__.py,sha256=2FfDaqwmJJCd-IVpY_CrzWcFvw0KFyQ3cm5jnTfI-DQ,771
-openai/resources/beta/threads/runs/runs.py,sha256=q3LJ5EaOwTIeuu2Li4JtH3HXj3WOcevL_7Lozj_1MOY,137198
-openai/resources/beta/threads/runs/steps.py,sha256=uRykb4JapSNZCF8OD54f5qOWtrp2GoU1k5uAZgA4kAk,12223
-openai/resources/beta/threads/threads.py,sha256=943MpBSJUcyf46KcAvE-JUIdmcausRd_oceYFAZ0GgE,93208
+openai/resources/beta/threads/runs/runs.py,sha256=7sPjaxa8Th6aXDeils1G8VKA9_2wsyjGUs5kJh3M50I,142593
+openai/resources/beta/threads/runs/steps.py,sha256=VlGD9NXtNqOt3uwlnepCavW7v3uVlvvyi0X1h9WZ_-E,15817
+openai/resources/beta/threads/threads.py,sha256=qGh4H0-42NhJHwPpyAYZlGx1ZgssFARJ45fhEDCyDQU,94238
 openai/resources/beta/vector_stores/__init__.py,sha256=11Xn1vhgndWiI0defJHv31vmbtbDgh2GwZT3gX8GgHk,1296
-openai/resources/beta/vector_stores/file_batches.py,sha256=rQvUKeBmTxMFYmUN1s67rqhKk_2TtmVtrCnz-urevtM,30849
-openai/resources/beta/vector_stores/files.py,sha256=wKWpbUIMLdAcTsMIprG7pVlQ1DFdA924AxWsYhm1S34,28572
-openai/resources/beta/vector_stores/vector_stores.py,sha256=DTzqhJ73TfNwVmfHZwZFz-qZdAzqPd2v4YUptUmOdhI,27809
+openai/resources/beta/vector_stores/file_batches.py,sha256=EomxymvX4oCIRXUAfKGShAYWqnv1vlAahcp_Wa7Kt7Y,31985
+openai/resources/beta/vector_stores/files.py,sha256=LjN6Zazb4dGV-xeQ-XRKAVciXsFj7LXh90AKJgVQ-Cw,29724
+openai/resources/beta/vector_stores/vector_stores.py,sha256=OnzaEjKov8npQQf9YSYljPOTNBzjfwmxfW_D7f7fLkQ,28916
 openai/resources/chat/__init__.py,sha256=8Q9ODRo1wIpFa34VaNwuaWFmxqFxagDtUhIAkQNvxEU,849
-openai/resources/chat/chat.py,sha256=Edexhbq1anfSS_I0wNRQb7rx1OV6-rq4sxgVlYDGb6Y,2342
-openai/resources/chat/completions.py,sha256=ZDneiCx57nohjlC3ENOuGrNU27JUZ0vNn7rhpzYekJk,76132
-openai/resources/completions.py,sha256=4Rfv9o3XwI5GRfhN1RD4tEgNn0I2jb6TRW6j0b6bpZc,58712
-openai/resources/embeddings.py,sha256=cMSXtMc_7mBqlSiQ99B7qXYoRLGyoeIFazyYQ0jJ1O4,10755
-openai/resources/files.py,sha256=InC0e28vgMKM5pNadUhGGvDQ0Rvb1D4WcEDfanFJV2E,27156
+openai/resources/chat/chat.py,sha256=hvYn24it5ARq8BYloSWn5kqqSlBEcYvVdQTf3ujxuV0,3360
+openai/resources/chat/completions.py,sha256=VL61UVRPoI7JuNj6b4k4G2g8Ew0mu2WfLJbtUbW_XuM,99603
+openai/resources/completions.py,sha256=5W3UuTH0V-vpTIkb8-r7gyS0Qp7tx3JZMWZkHBGIjPY,59460
+openai/resources/embeddings.py,sha256=PfwI3PKKPkmLs7wHijO-1pOwW6Fjs5Rqzpy0ALLYgAs,11655
+openai/resources/files.py,sha256=PL7b1lM7s3uJD7CvZcM_9f54kAlhBo913o31z1uXt-0,30093
 openai/resources/fine_tuning/__init__.py,sha256=s6uoq7gM4gwoywdOOZQkPeYiSbUl-OwpeuMhwJJk0lc,837
-openai/resources/fine_tuning/fine_tuning.py,sha256=-2k4d5ZDlCIoqonSOMtGLVl4Kk9n2yJoKvVMG3PoWW8,2410
+openai/resources/fine_tuning/fine_tuning.py,sha256=yfXXcR8IMRHkS-xnoT_nF7WEa2fjprDO-0ND-juPqhk,3394
 openai/resources/fine_tuning/jobs/__init__.py,sha256=_smlrwijZOCcsDWqKnofLxQM2QLucZzXgboL9zJBPHw,849
-openai/resources/fine_tuning/jobs/checkpoints.py,sha256=6uP1CCGkzE_n8FsVdTQ36eH_eiq24wOxQQ5zzOy0UEU,6456
-openai/resources/fine_tuning/jobs/jobs.py,sha256=6GOXP6GTRlmYeDaVA_FK0QDioudPtty7qNH4uo-_j58,27502
-openai/resources/images.py,sha256=1oKEnEalLjSxnyOe-Zzk3CsL9ou876ePUgOh8vtUc3I,24616
-openai/resources/models.py,sha256=XF3E56V62YZq-HrStUDDvfrT2RHj98P8Y-oOrPSPRX0,10222
-openai/resources/moderations.py,sha256=Jc6m5wsaWcqx9ls8HORqwqKZ-a7xy4mSpD3R3J-n7nc,6603
+openai/resources/fine_tuning/jobs/checkpoints.py,sha256=LIJUhxb8hgxEgHdTFKdyb0Q-hnV4ccIprvFpQJI97ho,7474
+openai/resources/fine_tuning/jobs/jobs.py,sha256=kZLZaWRW6ynhLknoOaK64LW9XifzsSOpFHWX8VPjJcs,29392
+openai/resources/images.py,sha256=PS7PIe1X8tccsqLtd-4kx1OTzCow0S-C-L29bmVyV4c,25634
+openai/resources/models.py,sha256=qJj0Cpy_Ok9ELag8VxqTefX8tw7RPgIZ8-a6qllxl8w,11240
+openai/resources/moderations.py,sha256=H9tygVKuT1c25LW_XyrhpK9nlT72SsEYDiPolQBP7hs,7805
 openai/resources/uploads/__init__.py,sha256=HmY3WQgvUI2bN3CjfWHWQOk7UUC6Ozna97_lHhrrRSA,810
-openai/resources/uploads/parts.py,sha256=8xBjww6Na7qx6QVTG-lyuKzgF1gs7ldP_VwnwMFFjd8,7132
-openai/resources/uploads/uploads.py,sha256=OpU8cdmgqm086JeDQ_Ck0_wiIFmpclTHwdMshyTjRP8,17618
-openai/types/__init__.py,sha256=vXLJXHVOMJwdhEZceBW9O6msKWH6Hmvs6G6XmN8MULQ,2571
-openai/types/audio/__init__.py,sha256=7PRM0dwBGCBKcF_hkDHx2tVUdEW5jbwU8bFB9Vbtq-g,547
-openai/types/audio/speech_create_params.py,sha256=Q7EqgD5F5CV0tANvz30msMfYD4EgqGUZn4V4yDypSe4,1300
+openai/resources/uploads/parts.py,sha256=NEMRVCqOOYJV2zTmBau9UtY2qXuB_yDJzzXTJ1XubUY,8150
+openai/resources/uploads/uploads.py,sha256=ft7cVZuDxphjdCV6BcS6Zs2qE3zD1RB57udvaGUR9HY,24918
+openai/types/__init__.py,sha256=GxEEa9qy8CKZVCU1wY4PokDUCq-fD_GwZxRsBxzC_-s,3177
+openai/types/audio/__init__.py,sha256=sR9_rMb-gO0stG4ozTq6XJs714C_BfjB3KCgFvyhXVA,1050
+openai/types/audio/speech_create_params.py,sha256=-iUZ3a-BGlg46IFsP_vcJBTRuK_pXruF0KJsbNn0mgU,1300
 openai/types/audio/speech_model.py,sha256=RUimvc__LYAxwEEmfrf-lj18O3EWrU1OlWZXEXN2AKY,218
 openai/types/audio/transcription.py,sha256=FP9QMwwwdqgvP3xY9P-40gBiFmMwFKxXM5yv5x8xPVk,230
-openai/types/audio/transcription_create_params.py,sha256=5tx0yAERDRZTG0IEsHQODKxMGZKrdRXGo5K2Is0gNw0,2253
+openai/types/audio/transcription_create_params.py,sha256=OP8fXaYYsi5HWi0E7HR5HIRihglsuBqeJWglxkNxLts,2264
+openai/types/audio/transcription_create_response.py,sha256=-PLGH8he9EdJtvBXV-ZrE31CLVnk4bc0VQ1ixRoN8Ck,378
+openai/types/audio/transcription_segment.py,sha256=-pPAGolwIIXUBMic-H5U7aR0u_Aq-pipSA4xTtn_viA,1153
+openai/types/audio/transcription_verbose.py,sha256=tlVK8JzyvkslQOvpAb19PmsfiRBqmbne0l-GqFmVIMU,758
+openai/types/audio/transcription_word.py,sha256=sNDdtjoqIiba6qKsD_lI2Ffs1Lr7qP9HyS59AFh5cTc,368
 openai/types/audio/translation.py,sha256=5l-Zk9Cg7AZti-TTn2-4ydsoZj2zdvDwyzzVjVp9W0g,194
-openai/types/audio/translation_create_params.py,sha256=GT1rk1U7nqInbyahHxBXX8uSjmXGCySiIhI53DYgpa4,1524
+openai/types/audio/translation_create_params.py,sha256=lFQEh5IRG5XT-Z3TV7FDSNbIRqAt6yA3EsSvSsb0wsU,1585
+openai/types/audio/translation_create_response.py,sha256=x6H0yjTbZR3vd3d7LdABcn9nrMDNdeMjepcjW1oUfVc,362
+openai/types/audio/translation_verbose.py,sha256=ic6h7_fAKlyrJuCgbd4Vtr0pk9OnynQK_uobD9lAGZo,613
 openai/types/audio_model.py,sha256=pxBVwf1HGd6mW-_jd-TDVMRZtTvvCUn_rL8Pt1BXzuo,208
-openai/types/batch.py,sha256=eIOIaJnDuv93fdefTI0WRfTm7MZH8gLBdF0B12JCiZw,2787
-openai/types/batch_create_params.py,sha256=Wq-uHe9FcAPTtN68jEG2xMZWwOC8Q7Dg4GdxV_y5qP0,1441
+openai/types/audio_response_format.py,sha256=EEItnQdwXinG8bOe1We2039Z7lp2Z8wSXXvTlFlkXzM,259
+openai/types/batch.py,sha256=Dq7btfgIT4b2yfh0knZTzAL4yFx_l95H5KLfDPO8iig,2788
+openai/types/batch_create_params.py,sha256=VXpg3mK2xwsUAIbYcFHFgRgLMrN3iBgW8l5rslk0gvQ,1441
 openai/types/batch_error.py,sha256=Xxl-gYm0jerpYyI-mKSSVxRMQRubkoLUiOP9U3v72EM,622
 openai/types/batch_list_params.py,sha256=X1_sfRspuIMSDyXWVh0YnJ9vJLeOOH66TrvgEHueC84,705
 openai/types/batch_request_counts.py,sha256=GHHrJKdJwJ3foBa1j9v5Vece_zzkdXXXgOcne8W1E30,409
-openai/types/beta/__init__.py,sha256=5ojZzNm9el-L9sXfh0E8D2t7N55jmuK_GMEkx9Yn09s,2723
-openai/types/beta/assistant.py,sha256=hn9R5DZ_WSwHRkFAX5biZGC48rMK8ZOf4tSn4J70AAs,4950
-openai/types/beta/assistant_create_params.py,sha256=Av6Cbm37NNMQyhhdhilmbFbrj329sipMiUYGcaFBlx0,7267
+openai/types/beta/__init__.py,sha256=CbOOxDPXvdK5RInCcEiBihJ2XgaUhdm3NMBBwx90OHc,3462
+openai/types/beta/assistant.py,sha256=3w8FpWceagZoKuEQrGeitoosTrz-Z24IPiL-viWC4I4,4936
+openai/types/beta/assistant_create_params.py,sha256=Y5LoiGU9ZTWQ87KaYyrqN1TsMFT4iYsBvMNeDgciRd4,5986
 openai/types/beta/assistant_deleted.py,sha256=bTTUl5FPHTBI5nRm7d0sGuR9VCSBDZ-IbOn9G_IpmJQ,301
-openai/types/beta/assistant_list_params.py,sha256=1-osjSX8tKieHSP0xaKBBU8j-J01fKrrxIJRHDudFHk,1220
+openai/types/beta/assistant_list_params.py,sha256=yW-lj6AUkG0IRZQKre0veEr9p4VMN-9YdELFMYs74Cw,1222
 openai/types/beta/assistant_response_format_option.py,sha256=yNeoAWxM-_8Sjmwqu8exqyKRFhVZIKeTypetPY55VFA,561
-openai/types/beta/assistant_response_format_option_param.py,sha256=t6eUp_Gogo-5_KOidEe2wKuZxVs72f9yp_HLc5rsDyA,488
-openai/types/beta/assistant_stream_event.py,sha256=ORGXB7viddEHvK4Nb40wqVJylWLgkwVXH7qlyYG9nQE,6829
+openai/types/beta/assistant_response_format_option_param.py,sha256=dyPMhwRSLBZ0ltpxiD7KM-9X6BzWnbGeG-nT_3SenuQ,628
+openai/types/beta/assistant_stream_event.py,sha256=vP4LDqYWzSKGcZ1JAfyNw7YqC__XsVPe0nqZ2qdn93E,6930
 openai/types/beta/assistant_tool.py,sha256=_0FC7Db4Ctq_0yLaKJ93zNTB5HthuJWEAHx3fadDRlw,506
 openai/types/beta/assistant_tool_choice.py,sha256=Hy4HIfPQCkWD8VruHHicuTkomNwljGHviQHk36prKhg,544
 openai/types/beta/assistant_tool_choice_function.py,sha256=aYMlVrZdX2JxmehDlyGALRK2PIEkO7VFEfsvY3VH6T4,270
@@ -449,19 +552,87 @@ openai/types/beta/assistant_tool_choice_option.py,sha256=jrXMd_IYIQ1pt8Lkc-KrPd4
 openai/types/beta/assistant_tool_choice_option_param.py,sha256=VcatO5Nej9e5eqfrwetG4uM1vFoewnBEcFz47IxAK2E,424
 openai/types/beta/assistant_tool_choice_param.py,sha256=NOWx9SzZEwYaHeAyFZTQlG3pmogMNXzjPJDGQUlbv7Q,572
 openai/types/beta/assistant_tool_param.py,sha256=6DcaU3nMjurur2VkVIYcCaRAY1QLQscXXjCd0ZHHGho,501
-openai/types/beta/assistant_update_params.py,sha256=6Eo_HUAJdAwRo7X-zzp4z8PVs9glPS-UV_EWO7aqZL8,4698
+openai/types/beta/assistant_update_params.py,sha256=XsLdjYNx7dbPr1aqDu0_ZGuXjgU0JVuM0waJo1NskyI,4684
+openai/types/beta/auto_file_chunking_strategy_param.py,sha256=hbBtARkJXSJE7_4RqC-ZR3NiztUp9S4WuG3s3W0GpqY,351
 openai/types/beta/chat/__init__.py,sha256=OKfJYcKb4NObdiRObqJV_dOyDQ8feXekDUge2o_4pXQ,122
 openai/types/beta/code_interpreter_tool.py,sha256=7mgQc9OtD_ZUnZeNhoobMFcmmvtZPFCNYGB-PEnNnfs,333
 openai/types/beta/code_interpreter_tool_param.py,sha256=X6mwzFyZx1RCKEYbBCPs4kh_tZkxFxydPMK4yFNJkLs,389
-openai/types/beta/file_search_tool.py,sha256=6OH6Vt9rV9Ym4U4G61PP6UdnfE-lMWg_HzBFW6bQBNc,974
-openai/types/beta/file_search_tool_param.py,sha256=iDUCeoUJ1uLvUI9rDwerXwDOULw9_bnCZgjXwfWlDi0,981
+openai/types/beta/file_chunking_strategy.py,sha256=6nRvYetBl_BHgN8biTyTut-tw8G13YttgxSKtJsJLeM,560
+openai/types/beta/file_chunking_strategy_param.py,sha256=P0x4I2hB_ylbSxFFEmRqgwto3HQQsHIokX3U0is_a9s,498
+openai/types/beta/file_search_tool.py,sha256=5aNU8RZj-UNdmuqqpjCXNaa1pI9GzSP5qCPtvVSJ1oQ,1769
+openai/types/beta/file_search_tool_param.py,sha256=o6sWPrzRYY8wtNaVuF8h3D1sAQV3N0L3dbdiiaMisW0,1765
 openai/types/beta/function_tool.py,sha256=oYGJfcfPpUohKw2ikgshDjOI1HXCK-5pAWyegYNezeU,397
-openai/types/beta/function_tool_param.py,sha256=T_k2OX1OULgkrHHXw0rY_J-O0y5qA0lM-B58C64YyfM,453
-openai/types/beta/thread.py,sha256=wd00j3ogUpOa_O0Sf1m6H4f8t1Nf05DKWiK_4m33O6s,2013
-openai/types/beta/thread_create_and_run_params.py,sha256=ZvuTJnslOhwzzEy946ICkftcopChwylK_KKLEHF3I3w,14477
-openai/types/beta/thread_create_params.py,sha256=0MlmA3nacpq1k7I-gxKvMes0Yo683grly1St4qUlOpQ,6215
+openai/types/beta/function_tool_param.py,sha256=hCclpGO4Re-TxiGy_QxX75g1kcN6_ElubicO6SdJ_YI,471
+openai/types/beta/other_file_chunking_strategy_object.py,sha256=hJz1OeSkvvcWJVftPfvz2pB5ujdawWEEa3v38E6tt7g,311
+openai/types/beta/realtime/__init__.py,sha256=OJOsvJMLlDqJEJClien1XwN8K6vhnyVtNgN1qolZeW0,6167
+openai/types/beta/realtime/conversation_created_event.py,sha256=U4-nesN8rAep2_25E2DrkXUMafQejj3NE_0llXKj5Y8,752
+openai/types/beta/realtime/conversation_item.py,sha256=av6WCjWVuRxBjccmxv4j26cd3TCKURj2a7cf8uS3P3s,2297
+openai/types/beta/realtime/conversation_item_content.py,sha256=S3Gqx5VBji5S64d0Rqq-r4HbUddSl_CT0xLZRfAjgmQ,930
+openai/types/beta/realtime/conversation_item_content_param.py,sha256=bIS_29Z_odnBth8as5pDmUFf1UNwJ9gfFCYlz8ZdRRQ,852
+openai/types/beta/realtime/conversation_item_create_event.py,sha256=PNdOLjWMB2uc0tCm7QdWANXt7FWqKpgocnej2OiEjxw,976
+openai/types/beta/realtime/conversation_item_create_event_param.py,sha256=L9e8U-3LITXlBuJ_FQfGhSDX3Jj7R3uWN1UiG7qDTec,996
+openai/types/beta/realtime/conversation_item_created_event.py,sha256=DIeG7YQ5HdKrnbnorklB1Zfsz42yRdPKDOx5TPzfvw0,722
+openai/types/beta/realtime/conversation_item_delete_event.py,sha256=p-O6R1Ku5pxZvaxhSi4YTPqLXS1SHhdLGgJuPQyPcHY,549
+openai/types/beta/realtime/conversation_item_delete_event_param.py,sha256=a17h8Hd8MxUbXT6NQg8YpTr1ICt1ztRecpfukHw4g34,569
+openai/types/beta/realtime/conversation_item_deleted_event.py,sha256=uWHSqX5ig550romSdhtROwrdQmdeN31Oz1Vpr9IuQFI,492
+openai/types/beta/realtime/conversation_item_input_audio_transcription_completed_event.py,sha256=7tX1hI3g0SbrXGHcaC_Y1xAzhsoziReYwlqyA8ycB3E,764
+openai/types/beta/realtime/conversation_item_input_audio_transcription_failed_event.py,sha256=xYNSBIyERQJ4P-5YoFF1VptfPa8JnJ0sWaH6LGsPow0,1077
+openai/types/beta/realtime/conversation_item_param.py,sha256=x12A5-yjNWodFNJEnbHKY1WJzSzX9s7EQr2c5FuYKBQ,2177
+openai/types/beta/realtime/conversation_item_truncate_event.py,sha256=1c2_BamaTkgD26eyGZJU5xwbz7lRHupqU2HqcK0VniI,943
+openai/types/beta/realtime/conversation_item_truncate_event_param.py,sha256=hSnVOSMMtLf16nn4ISHkevYCfEsiN9kNcgxXRtHa8Kc,983
+openai/types/beta/realtime/conversation_item_truncated_event.py,sha256=K4S35U85J-UNRba9nkm-7G1ReZu8gA8Sa1z0-Vlozc0,704
+openai/types/beta/realtime/error_event.py,sha256=goNkorKXUHKiYVsVunEsnaRa6_3dsDKVtrxXQtzZCmk,877
+openai/types/beta/realtime/input_audio_buffer_append_event.py,sha256=lTKWd_WFbtDAy6AdaCjeQYBV0dgHuVNNt_PbrtPB8tg,662
+openai/types/beta/realtime/input_audio_buffer_append_event_param.py,sha256=XmN2bE6jBRrkKGVPJdnPjJql5dqMPqwbmFnxo-z22JE,682
+openai/types/beta/realtime/input_audio_buffer_clear_event.py,sha256=7AfCQfMxZQ-UoQXF9edYKw5GcTELPcfvvJWWpuLS41c,489
+openai/types/beta/realtime/input_audio_buffer_clear_event_param.py,sha256=y-zfWqJsh1n6r2i0MgLDpnNC4g1dq3GCS66Twfkng38,499
+openai/types/beta/realtime/input_audio_buffer_cleared_event.py,sha256=j9gpm7aGVmrUt48wqtvBMN8NOgtvqHciegjXjOnWm7A,429
+openai/types/beta/realtime/input_audio_buffer_commit_event.py,sha256=SLZR2xxRd6uO3IQL6-LuozkjROXiGyblKoHYQjwXk4I,493
+openai/types/beta/realtime/input_audio_buffer_commit_event_param.py,sha256=B8agXC-rUl-D-RijJ5MeTLgw43qVYzmf2_2oAVokhLY,503
+openai/types/beta/realtime/input_audio_buffer_committed_event.py,sha256=wXMxuXLw1jmT4e-FmTp6rSxcSc_4l55zO3gT7jI1Mp4,628
+openai/types/beta/realtime/input_audio_buffer_speech_started_event.py,sha256=NVp60RUsLFtte9Ilknmu_5lRk2dZp_1fXCgGHd4EvSM,861
+openai/types/beta/realtime/input_audio_buffer_speech_stopped_event.py,sha256=gszRuYQtAW8upIhd7CJZ7pxboDk-K7sqidjqxgf47q4,779
+openai/types/beta/realtime/rate_limits_updated_event.py,sha256=kBnf_p-49Q_LNdJsj0R1Szi8R4TGYAAJ_KifLuuyFZw,949
+openai/types/beta/realtime/realtime_client_event.py,sha256=TD_qJi1hNgvurWTUzG-xb27thuvUT2-2AK_pouAY3vc,1249
+openai/types/beta/realtime/realtime_client_event_param.py,sha256=qNStVbW_imzF0F8qfEHHE07AZoPIQLvjcTw9mXu4mFY,1294
+openai/types/beta/realtime/realtime_connect_params.py,sha256=AvTypkFCYmDn9qMeektVqij6cqzgovr3PpgpMalJoJ4,290
+openai/types/beta/realtime/realtime_response.py,sha256=C-3ZTF_gy40eT1eaeWIfpBS3pQC5lv3XNM_mqiLtTWg,1505
+openai/types/beta/realtime/realtime_response_status.py,sha256=gU-59Pr_58TRfMZqFzdCloc53e1qOnU4aaHY3yURUK8,1326
+openai/types/beta/realtime/realtime_response_usage.py,sha256=6XOFjCjPWioHoICZ0Q8KXuUzktQugx6WuTz0O5UvzZg,1541
+openai/types/beta/realtime/realtime_server_event.py,sha256=j8s9jdl5cARv3fVM5jEjo04f83FmNELPRS_lq5Ao_Q0,3512
+openai/types/beta/realtime/response_audio_delta_event.py,sha256=UjbnK4u_WSNTOColZj8SmJgHnAc2H8iRXD76ZnPbz7E,742
+openai/types/beta/realtime/response_audio_done_event.py,sha256=1XEWBPh1JiOgyr6V03mRt_3sLm0YFUq5ft1AhfFlNEg,679
+openai/types/beta/realtime/response_audio_transcript_delta_event.py,sha256=HEVNQ_R2_Nyo6BvNvsliMnN__b17eVd2Jx5udRHg0Hg,773
+openai/types/beta/realtime/response_audio_transcript_done_event.py,sha256=Cn5l4mJnKK3LeSN9qFL4LLqs1WOWg4kt1SaYThB-5c0,787
+openai/types/beta/realtime/response_cancel_event.py,sha256=EKx8IZUISJHdl-_3tCdHtz2BINQ85Tq_ocadnsEGPSk,637
+openai/types/beta/realtime/response_cancel_event_param.py,sha256=nidzBL83liHwyImiNGiz9Ad0V34EtFAQDw1utqcF6ns,630
+openai/types/beta/realtime/response_content_part_added_event.py,sha256=a8-rm1NAwX685fk7GdT6Xi0Yr-JfeAkyUr94-RoFe34,1232
+openai/types/beta/realtime/response_content_part_done_event.py,sha256=jO2TZygxPabbnEG9E1AfNP-JYJv1QtCMnCzgcZ_3n18,1190
+openai/types/beta/realtime/response_create_event.py,sha256=a6oMZOu4htnwQhrroU5LOXqDpbWMvK7PZpznlQcgkqs,4423
+openai/types/beta/realtime/response_create_event_param.py,sha256=jRpgQznlLqgQrR3QyBhv8x1tZVUJaF3t549eaankq5c,4268
+openai/types/beta/realtime/response_created_event.py,sha256=zZtHx-1YjehXxX6aNE88SFINDaKOBzpzejo6sTNjq9g,506
+openai/types/beta/realtime/response_done_event.py,sha256=_yUPoECCli89iHLtV3NQkXQOW6Lc1JlxVPFw04ziBGY,494
+openai/types/beta/realtime/response_function_call_arguments_delta_event.py,sha256=Yh2mQZDucfnTLiO8LRyG9r7zeS1sjwLcMF1JPMdTFJc,793
+openai/types/beta/realtime/response_function_call_arguments_done_event.py,sha256=kxSPK6nbNWL6pxveY7zaNGgCkCXqyBFJPVYJrw9cbOw,793
+openai/types/beta/realtime/response_output_item_added_event.py,sha256=-_BZjvAqcgv3NIz-EMhvYMxIwvcXTt68FVNp0pw09dI,713
+openai/types/beta/realtime/response_output_item_done_event.py,sha256=0ClNVMZmeIxKghlEid9VGoWiZ97wp00hIdNnev4qBD8,709
+openai/types/beta/realtime/response_text_delta_event.py,sha256=B1yyuc6iMOMoG5Wh6W5KoQNYtVD1vEm2cKqHnl2CuFQ,721
+openai/types/beta/realtime/response_text_done_event.py,sha256=mPgVG6nWxwkZ3aZOX-JkVF7CpaWP5-bvtbxFrr4fK7g,724
+openai/types/beta/realtime/session.py,sha256=OqdK0L7ugOYV0PT2XlixRERimneHIF7-oHUh1JWtK70,5388
+openai/types/beta/realtime/session_create_params.py,sha256=ALVf2hYtcdZjl2A5LJyjiyPfqSFEAeTIVkELbeSaH-g,5161
+openai/types/beta/realtime/session_create_response.py,sha256=iyovJfORab-aDJJKE8PN--VQeCBR3VlnyV1tf4qE-K0,5411
+openai/types/beta/realtime/session_created_event.py,sha256=rTElnBlE7z1htmkdmpdPN4q_dUYS6Su4BkmsqO65hUc,489
+openai/types/beta/realtime/session_update_event.py,sha256=VwRvNgu-otI5_0xnXso1gqlCEWFnqzrGq9-Kar_o71Q,5751
+openai/types/beta/realtime/session_update_event_param.py,sha256=NmDaFhVTohrHi-yRd1x883NUjGH3N6ZWyfpfJ0tEpTQ,5573
+openai/types/beta/realtime/session_updated_event.py,sha256=HyR-Pz3U9finVO-bUCvnmeqsANw-fceNvVqEIF6ey10,489
+openai/types/beta/static_file_chunking_strategy.py,sha256=nHaLv70q1rencY2u8mqS7mW7X7enzHrc-zM9mg22dHw,597
+openai/types/beta/static_file_chunking_strategy_object.py,sha256=aOPxudte299F0j3bzniXcKJ7j-w4ZfQpgFHTa3CFyZ8,425
+openai/types/beta/static_file_chunking_strategy_param.py,sha256=kCMmgyOxO0XIF2wjCWjUXtyn9S6q_7mNmyUCauqrjsg,692
+openai/types/beta/thread.py,sha256=9wxx6M26S7cilx5SKWjZnkHc7g222AIOhikd0WTJfwI,2014
+openai/types/beta/thread_create_and_run_params.py,sha256=NHkj-IMm2WEqH82i9zxqgJqYkOVCBVXSpZcpl-SVznY,13175
+openai/types/beta/thread_create_params.py,sha256=U0gNXfSltPqYF3GIGQ7dloolkz6nzuDimXF-V9wjzvo,4970
 openai/types/beta/thread_deleted.py,sha256=MaYG_jZIjSiB9h_ZBiTtpMsRSwFKkCY83ziM5GO_oUk,292
-openai/types/beta/thread_update_params.py,sha256=RYsR88YHwReKLiLqnLlnWiReiVIGlEGvVV9-g_wptgM,1750
+openai/types/beta/thread_update_params.py,sha256=olIjwn1eD0H2AkjdDZC38lPdT5dp2ORSjavPA7pB_08,1751
 openai/types/beta/threads/__init__.py,sha256=0WsJo0tXp08CgayozR7Tqc3b8sqzotWzvBun19CEIWc,3066
 openai/types/beta/threads/annotation.py,sha256=Ce3Y0mSodmYRkoqyhtyIdep6WfWew6KJJgtrENOnfek,462
 openai/types/beta/threads/annotation_delta.py,sha256=iNsE-1Gn1yU0TlTHoxqKbOvPRUxWuXsF72qY_mMnWGY,510
@@ -481,40 +652,42 @@ openai/types/beta/threads/image_url_content_block_param.py,sha256=RWzo5KkBiwvgJS
 openai/types/beta/threads/image_url_delta.py,sha256=MXCp-OmuNT4njbWA9DWAbocP7pD3VpdcUy2wgeOjwm4,582
 openai/types/beta/threads/image_url_delta_block.py,sha256=Jjdfub4g9ceNKF8GuuTIghOmYba2vEeX3320mg5PWIA,484
 openai/types/beta/threads/image_url_param.py,sha256=VRLaxZf-wxnvAOcKGwyF_o6KEvwktBfE3B6KmYE5LZo,602
-openai/types/beta/threads/message.py,sha256=r3Lj8coZqEoz-2nT11yivi7K25CcUskbNSo4x7sWXPg,3294
+openai/types/beta/threads/message.py,sha256=aGWe0kiNv5sXUYheJ0o1KpTds4oTaeDmqot1PMStJCE,3295
 openai/types/beta/threads/message_content.py,sha256=b8IC_EG28hcXk28z09EABfJwPkYZ7U-lTp_9ykdoxvU,630
 openai/types/beta/threads/message_content_delta.py,sha256=o4Edlx9BtdH2Z4OMwGWWXex8wiijknNRihJ-wu8PDUQ,615
 openai/types/beta/threads/message_content_part_param.py,sha256=RXrnoDP2-UMQHoR2jJvaT3JHrCeffLi6WzXzH05cDGI,550
-openai/types/beta/threads/message_create_params.py,sha256=Qs7Gxs8ZKwzk_7ZhJOwj4KiHkidSmy5qc_Dam0P8F5E,1956
+openai/types/beta/threads/message_create_params.py,sha256=WYfc_-kc7lxcxdpwKCVT2Ei-5Jl_132uqOHMtXL92OE,1957
 openai/types/beta/threads/message_deleted.py,sha256=DNnrSfGZ3kWEazmo4mVTdLhiKlIHxs-D8Ef5sNdHY1o,303
 openai/types/beta/threads/message_delta.py,sha256=-kaRyvnIA8Yr2QV5jKRn15BU2Ni068a_WtWJ4PqlLfE,570
 openai/types/beta/threads/message_delta_event.py,sha256=7SpE4Dd3Lrc_cm97SzBwZzGGhfLqiFViDeTRQz-5YmQ,579
-openai/types/beta/threads/message_list_params.py,sha256=LXqc3deSkKO6VN337OlQ4fzG7dfgBE7Iv_CLzZHhbhw,1294
-openai/types/beta/threads/message_update_params.py,sha256=bw6_U-vZA4c9_CDmeGOh7IEPIm8BU3BBOKtxnii0LKA,629
+openai/types/beta/threads/message_list_params.py,sha256=iuwzDccnViooUxHlq-WoE1FEJArNy5-zrYCoaNgVS8k,1296
+openai/types/beta/threads/message_update_params.py,sha256=jTM_WDKDuPVJKlNKlT6J_UqQjgM2vrrD03ZhvHI5bSY,630
 openai/types/beta/threads/refusal_content_block.py,sha256=qB9jrS2Wv9UQ7XXaIVKe62dTAU1WOnN3qenR_E43mhg,310
 openai/types/beta/threads/refusal_delta_block.py,sha256=ZhgFC8KqA9LIwo_CQIX-w3VVg3Vj0h71xC1Hh1bwmnU,423
 openai/types/beta/threads/required_action_function_tool_call.py,sha256=XsR4OBbxI-RWteLvhcLEDBan6eUUGvhLORFRKjPbsLg,888
-openai/types/beta/threads/run.py,sha256=_yc01xJzML8ekoB5mX9HGxlJUJmPaiukg0rpuQAK3Rc,8211
-openai/types/beta/threads/run_create_params.py,sha256=4wxYnpiZfeUfkW79Y3_qLJgqEHnY2b1_SXvyLwaD_FQ,9163
-openai/types/beta/threads/run_list_params.py,sha256=73poqeRcb5TEsIVn7OzJ_g9OajNokEzpCVLzVNKZmPk,1208
+openai/types/beta/threads/run.py,sha256=GR469hvbAlWTHL17MieCYxQfASyxaY1ZOe6Qbf0ORMI,8218
+openai/types/beta/threads/run_create_params.py,sha256=KgltVibs_KnKsL3UaZyVJgb-6aUxct7CXUtqMdkTXTM,9670
+openai/types/beta/threads/run_list_params.py,sha256=TgepSLrupUUtuQV2kbVcoGH1YA0FVUX9ESkszKuwyHY,1210
 openai/types/beta/threads/run_status.py,sha256=OU1hzoyYXaRJ3lupX4YcZ-HZkTpctNE4tzAcp6X8Q9U,351
-openai/types/beta/threads/run_submit_tool_outputs_params.py,sha256=aDrg0FZZoJKaPVQzcFjUg4ZKaeW8KF6UJBxhJEIjC2I,1630
-openai/types/beta/threads/run_update_params.py,sha256=76dWMNa3zCUliemCdwWv6p07GNeMYCdZoJs9KNbdZSE,621
-openai/types/beta/threads/runs/__init__.py,sha256=uhxk5F1_5c5wg2_p70AjlOy9cE3Ga8-ILn4Ep-gcls4,1515
+openai/types/beta/threads/run_submit_tool_outputs_params.py,sha256=cKiyD374BsZN_Oih5o5n5gOf_DYsxErVrbgxveNhmPI,1643
+openai/types/beta/threads/run_update_params.py,sha256=EDYJO3YuH1IKjfR1xAaBtWFonNnyXJDYAnlaMnwyXo8,622
+openai/types/beta/threads/runs/__init__.py,sha256=mg_roY9yL1bClJ8isizkQgHOAkN17iSdVr2m65iyBrs,1653
 openai/types/beta/threads/runs/code_interpreter_logs.py,sha256=7wXZpUE9I-oZJ0K3mFG0Nwmfm2bKGiSpWJyBeo7txwo,482
 openai/types/beta/threads/runs/code_interpreter_output_image.py,sha256=8o99k0ZHMHpqH0taXkOkYR9WaDUpCN-G0Ifd5XsJpb8,613
 openai/types/beta/threads/runs/code_interpreter_tool_call.py,sha256=ekiIuH1kVCN51hCzY3AYr5i3_a4vlgUiZHJ59pl17oY,1810
 openai/types/beta/threads/runs/code_interpreter_tool_call_delta.py,sha256=Qr2cen-bKyXTW2NDEUHnmJRE0jY-nkLcnO4NzCbBPDo,1479
-openai/types/beta/threads/runs/file_search_tool_call.py,sha256=PPxrJP3r4RWFTeE5mU-9SbFz37JmKHOGfsxlZGydyW0,522
+openai/types/beta/threads/runs/file_search_tool_call.py,sha256=XBgsM_USVr3ZrwTZx4L1-YG94Qv8c8GXI19ZHtDrZq8,1897
 openai/types/beta/threads/runs/file_search_tool_call_delta.py,sha256=Gx8c7GSgGYuOvGadcAr3ZIspEFMZS3e2OY7vBo_MYnM,655
 openai/types/beta/threads/runs/function_tool_call.py,sha256=aOq5yOtKOi6C5Q1FIQRxqtJJR1AcSW_K5PvRiKISNCI,920
 openai/types/beta/threads/runs/function_tool_call_delta.py,sha256=VFRtCJkj4PHX97upM1cXpJAk9-JvJSgyngie06fBIjQ,1076
 openai/types/beta/threads/runs/message_creation_step_details.py,sha256=tRFMNF2Rf4DekVliUKkoujItiOjjAE9EG9bbxJvpVPA,506
-openai/types/beta/threads/runs/run_step.py,sha256=NA3QgTsa646h3LYV13K3CdpsEQHaaD1QN2zT7STo6zo,3468
+openai/types/beta/threads/runs/run_step.py,sha256=L_CiwlW9y7NEOTumv1RyoQrQ_oCaNowRmraUHiAgJEc,3469
 openai/types/beta/threads/runs/run_step_delta.py,sha256=FNYDTddRrTO3PT_fgi7AsJ1PeMtyWsVzcxoihjbBzAw,663
 openai/types/beta/threads/runs/run_step_delta_event.py,sha256=rkDyvHSXt-hc1LngB41f9vglkn6t03kS62bsn0iGaxU,585
 openai/types/beta/threads/runs/run_step_delta_message_delta.py,sha256=UIo6oPH8STLjPHiWL-A4CtKfYe49uptvIAHWNnZ3Ums,564
-openai/types/beta/threads/runs/step_list_params.py,sha256=2vMPFMElvK135ncP9ch6kUnzPGOSIPT3Eio18jJhAqk,1250
+openai/types/beta/threads/runs/run_step_include.py,sha256=u-9Cw1hruRiWr70f_hw4XG0w1cwOAYfRJYKva2dEacs,264
+openai/types/beta/threads/runs/step_list_params.py,sha256=zorF5juogCzLMsZLjzMZTs_iIBcPj9WUur5HcrXuH8M,1752
+openai/types/beta/threads/runs/step_retrieve_params.py,sha256=aJ7l8RDJLPyEmqjfO4XsTV54VZOOqyb_gKSUvqp33ZI,815
 openai/types/beta/threads/runs/tool_call.py,sha256=1rwq4IbLgjQAQ-ORXYkNpmJyi9SREDnqA57nJbj_NiU,537
 openai/types/beta/threads/runs/tool_call_delta.py,sha256=t5wF8ndW3z99lHF981FL-IN5xXBS9p7eonH9bxvKu_c,600
 openai/types/beta/threads/runs/tool_call_delta_object.py,sha256=eK20VsIswEyT48XbkGu60HUrE7OD3fhpn1fbXrVauM4,615
@@ -524,82 +697,94 @@ openai/types/beta/threads/text_content_block.py,sha256=pdGlKYM1IF9PjTvxjxo1oDg1X
 openai/types/beta/threads/text_content_block_param.py,sha256=feQr0muF845tc1q3FJrzgYOhXeuKLU3x1x5DGFTN2Q0,407
 openai/types/beta/threads/text_delta.py,sha256=2EFeQCkg_cc8nYEJ6BtYAA3_TqgMTbmEXoMvLjzaB34,389
 openai/types/beta/threads/text_delta_block.py,sha256=pkHkVBgNsmHi9JURzs5ayPqxQXSkex3F0jH0MqJXik0,448
-openai/types/beta/vector_store.py,sha256=zaSaSUpStD3iuyas9f7VQCNF1byxnXRz_5q36eizNGE,2353
-openai/types/beta/vector_store_create_params.py,sha256=Q-pD0r2gg0PpTQdrltvQ-YdG8dDF48i8EoBsr0PRPWM,2509
+openai/types/beta/vector_store.py,sha256=R8M70uuGWVKt4t0ef__Py-MPw33Ljx4sh5ddihJMbIU,2354
+openai/types/beta/vector_store_create_params.py,sha256=rvvYUSDBbc5L6PAiMGSFQD85ugyR9mLdvZMxjap0fnk,1600
 openai/types/beta/vector_store_deleted.py,sha256=Yq0E1orRLShseLwZ1deiBdDEUgEw_tcYVxGYa5gbIrM,308
-openai/types/beta/vector_store_list_params.py,sha256=8iUgSgs_TeehprKjtTLWOGeH_R8LbDdLkdwMq9xVpSA,1224
-openai/types/beta/vector_store_update_params.py,sha256=AHlOV4f36UWAH4k7XKlGa51Mfao2f7339qI3fskWbIk,1114
+openai/types/beta/vector_store_list_params.py,sha256=KeSeQaEdqO2EiPEVtq1Nun-uRRdkfwW0P8aHeCmL5zA,1226
+openai/types/beta/vector_store_update_params.py,sha256=6OEP1IvilrGoPhHQPXOMQA0TwmCubeo7rB_ik5GQSrY,1115
 openai/types/beta/vector_stores/__init__.py,sha256=gXfm8V5Ad0iueaC_VoHDUQvSdwSfBzk2cQNwZldvY0s,671
-openai/types/beta/vector_stores/file_batch_create_params.py,sha256=c3Syo18qBuM1NYOopZN0CLigi864uo9jyFa4W6lH4i4,1922
-openai/types/beta/vector_stores/file_batch_list_files_params.py,sha256=6c_KvnlFV0vkFid_thhyEK6HC6F1ixbDh2roExL_-qk,1449
-openai/types/beta/vector_stores/file_create_params.py,sha256=6gCvIuEgvaPIIGS8IHwWQVTfwPgDN2Mt5zrDZY9z-4M,1890
-openai/types/beta/vector_stores/file_list_params.py,sha256=UC6NzZQ79tInL8xV3pMm66IFWsIT9PW_BhSbQLm4ar4,1383
-openai/types/beta/vector_stores/vector_store_file.py,sha256=TxefqQwU1vrJdINGdv91EQjyGZ-5eBqjAcUJAPvVETM,2730
+openai/types/beta/vector_stores/file_batch_create_params.py,sha256=lV4t5kikvEhl431RZgGDyQdFKTl-zXI-Q7YnbM0Qmv8,798
+openai/types/beta/vector_stores/file_batch_list_files_params.py,sha256=FPpQvCQI2skyLB8YCuwdCj7RbO9ba1UjaHAtvrWxAbs,1451
+openai/types/beta/vector_stores/file_create_params.py,sha256=kwSqe-le2UaYrcXGPxlP41QhH2OGvLXBbntAGlmK288,748
+openai/types/beta/vector_stores/file_list_params.py,sha256=AIzmNH1oFuy-qlpRhj9eXu9yyTA-2z_IppLYFclMtZw,1385
+openai/types/beta/vector_stores/vector_store_file.py,sha256=X8aQg4jYlK7iQumxn7B-eammIKVjUbu4lapPeq9jDWo,1788
 openai/types/beta/vector_stores/vector_store_file_batch.py,sha256=ubvj8z95EOdRGAp0rgI94g5uFQx0ob8hLgwOWHKda4E,1457
 openai/types/beta/vector_stores/vector_store_file_deleted.py,sha256=37J7oL2WYCgOd7Rhg2jX6IavaZT63vgUf3u6LC6C3Hs,322
-openai/types/chat/__init__.py,sha256=epD7g5z--KfkvxuhuvFS1uXFrlrV3djgoR-ORTYkbjI,3050
+openai/types/chat/__init__.py,sha256=coi_C98uX9XhThMVJ0GgjPVpzOYOMgj-ZmCWulEE3EA,3849
 openai/types/chat/chat_completion.py,sha256=MaTVOMwtbzqGyHgyP4DP41ESEDKhv_XOM8L_fx3uoQE,2689
-openai/types/chat/chat_completion_assistant_message_param.py,sha256=0m5WjA97DuxiGGvUyJQnlkf1SqLEr2Ce-kUTBvtLbBc,2114
+openai/types/chat/chat_completion_assistant_message_param.py,sha256=E6ZrsjEN_JHOHO-wC7Uk90Fa7Qz7bfgx8jea0z6g30s,2421
+openai/types/chat/chat_completion_audio.py,sha256=vzWeaAAAbomkvbFksXQu6qpw1RVJiuFytJZswO6h6vI,656
+openai/types/chat/chat_completion_audio_param.py,sha256=MnY4PNK8-OOaODkHNhBbSbzH4HmqykKvwftsOjVpOAE,801
 openai/types/chat/chat_completion_chunk.py,sha256=aQXFY4gq9YEIrr7YBM68D5XyWGT9kKo0JO8n-55IjEA,5032
-openai/types/chat/chat_completion_content_part_image_param.py,sha256=ODHcWpe8TIXZQHXHhEEacrRHm_TCaFWZnml-bD85XiU,797
-openai/types/chat/chat_completion_content_part_param.py,sha256=8hoTnNqerHjaHGMFU8CvhjVbH8yChXEYxs3jLWKfod8,543
+openai/types/chat/chat_completion_content_part_image_param.py,sha256=Gqv98qyD8jB81THZp49c8v2tHrId_iQp4NzciT9SKI0,797
+openai/types/chat/chat_completion_content_part_input_audio_param.py,sha256=r1EXNEtjJo5oJ9AnP3omaJzACE1gSfdmob5Q0HKsOm4,704
+openai/types/chat/chat_completion_content_part_param.py,sha256=7lCk-fZB5iT5keHLWw9eM-Hd5jsnPh2IIHICIUpoEXk,686
 openai/types/chat/chat_completion_content_part_refusal_param.py,sha256=TV1vu-IgrvKa5IBlPSIdBxUaW8g1zDhMOOBOEmhU2w0,467
 openai/types/chat/chat_completion_content_part_text_param.py,sha256=4IpiXMKM9AuTyop5PRptPBbBhh9s93xy2vjg4Yw6NIw,429
+openai/types/chat/chat_completion_developer_message_param.py,sha256=OCFKdTWkff94VtgY7AaDUUFiZLT8LBn7WWxjbcIq2OM,830
 openai/types/chat/chat_completion_function_call_option_param.py,sha256=M-IqWHyBLkvYBcwFxxp4ydCIxbPDaMlNl4bik9UoFd4,365
 openai/types/chat/chat_completion_function_message_param.py,sha256=jIaZbBHHbt4v4xHCIyvYtYLst_X4jOznRjYNcTf0MF0,591
-openai/types/chat/chat_completion_message.py,sha256=CYVebAMTUfREmvkykqXSNE6tGzEJu1QzClZ_ZgFD73s,1371
-openai/types/chat/chat_completion_message_param.py,sha256=RFer4ZYXxVed9F0ulkqi0xNy_eOhp63Y-0oN24dhVBI,889
+openai/types/chat/chat_completion_message.py,sha256=AH7JpjgKfphxBRJyI4PhwHCMREy_-D-a4_4u4NHjSfc,1674
+openai/types/chat/chat_completion_message_param.py,sha256=aLrz_cX_CYymFdW9cMIPZpv0Z4zM50RECV3SH6QNZsc,1019
 openai/types/chat/chat_completion_message_tool_call.py,sha256=XlIe2vhSYvrt8o8Yol5AQqnacI1xHqpEIV26G4oNrZY,900
 openai/types/chat/chat_completion_message_tool_call_param.py,sha256=XNhuUpGr5qwVTo0K8YavJwleHYSdwN_urK51eKlqC24,1009
+openai/types/chat/chat_completion_modality.py,sha256=8Ga0kruwJc43WD2OIqNudn7KrVRTPDQaalVkh_8bp9I,236
 openai/types/chat/chat_completion_named_tool_choice_param.py,sha256=JsxfSJYpOmF7zIreQ0JrXRSLp07OGCBSycRRcF6OZmg,569
+openai/types/chat/chat_completion_prediction_content_param.py,sha256=Xw4K_4F379LsXENOpZvREDn55cCnbmZ69xa4fw9w3bg,868
+openai/types/chat/chat_completion_reasoning_effort.py,sha256=Bs4xRaukXpM-_NW-QSKKnUyIPDw1ffSqnWaHU-rMdIE,258
 openai/types/chat/chat_completion_role.py,sha256=Rdzg4deI1uZmqgkwnMrLHvbV2fPRqKcHLQrVmKVk9Dw,262
 openai/types/chat/chat_completion_stream_options_param.py,sha256=7-R2mYh7dbtX9qDOL3UkeyVH6FNWC_4aTCLtHYObMbs,628
 openai/types/chat/chat_completion_system_message_param.py,sha256=WYtzmsNP8ZI3Ie8cd-oU7RuNoaBF6-bBR3mOzST9hMw,815
 openai/types/chat/chat_completion_token_logprob.py,sha256=6-ipUFfsXMf5L7FDFi127NaVkDtmEooVgGBF6Ts965A,1769
 openai/types/chat/chat_completion_tool_choice_option_param.py,sha256=ef71WSM9HMQhIQUocRgVJUVW-bSRwK2_1NjFSB5TPiI,472
 openai/types/chat/chat_completion_tool_message_param.py,sha256=5K7jfKpwTuKNi1PTFabq_LHH-7wun8CUsLDh90U8zQE,730
-openai/types/chat/chat_completion_tool_param.py,sha256=sve2G1DayUs-1CMzXK1x104r8KTa5K62CZdxoyLmFlk,485
+openai/types/chat/chat_completion_tool_param.py,sha256=J9r2TAWygkIBDInWEKx29gBE0wiCgc7HpXFyQhxSkAU,503
 openai/types/chat/chat_completion_user_message_param.py,sha256=mik-MRkwb543C5FSJ52LtTkeA2E_HdLUgtoHEdO73XQ,792
-openai/types/chat/completion_create_params.py,sha256=WT1UacPa7U_YlOgwClm_xBuoPdwApX-VRQmTsKFdwx4,11205
+openai/types/chat/completion_create_params.py,sha256=CGwTjckVhpxaQfA9zRKmrMCHvnYk-eaPFVmSVoA5Nls,13926
 openai/types/chat/parsed_chat_completion.py,sha256=KwcwCtj0yexl6gB7yuOnyETRW-uUvNRYbVzPMkwCe5Q,1437
 openai/types/chat/parsed_function_tool_call.py,sha256=hJzcKOpzf1tnXC6RGbPhaeCawq8EFdnLK_MfRITkW1U,920
-openai/types/chat_model.py,sha256=sR4hutPtTSFNrs3l5Rfj_lkDz_t11z1BjeBDKf9C4kg,755
+openai/types/chat_model.py,sha256=k9Ic_l5usRyY6xSHnqe4dBMKM5R4klTGuANg6z88WFk,1107
 openai/types/completion.py,sha256=yuYVEVkJcMVUINNLglkxOJqCx097HKCYFeJun3Js73A,1172
 openai/types/completion_choice.py,sha256=PUk77T3Cp34UJSXoMfSzTKGWDK0rQQwq84X_PSlOUJo,965
-openai/types/completion_create_params.py,sha256=mEyR068kk36ZylY4d1K3sqnucpUz9fAqEyoEwmW3DtQ,7567
-openai/types/completion_usage.py,sha256=9m5PmCyfVy93ZfIszGpcG1gdcEaSr22HJgaTJ_ImVrs,435
+openai/types/completion_create_params.py,sha256=TWNRWlGAcvirzY3Piy6AeYKyNxG7ktmtwjS27Q4bTi8,7535
+openai/types/completion_usage.py,sha256=uf5n0vzlCkGAU67BBn_h7yhjd_G4OHpQbJnvzz0eO2A,1735
 openai/types/create_embedding_response.py,sha256=lTAu_Pym76kFljDnnDRoDB2GNQSzWmwwlqf5ff7FNPM,798
 openai/types/embedding.py,sha256=2pV6RTSf5UV6E86Xeud5ZwmjQjMS93m_4LrQ0GN3fho,637
-openai/types/embedding_create_params.py,sha256=3p7U8i2uG1SCpELbn_IeDMLkFe-vv7cyB5dx-_4U8iU,1885
+openai/types/embedding_create_params.py,sha256=C9Tm1C_m96QtjyNc8fiy6wzs9HkM2GUF8CSTSS6V7ks,1850
+openai/types/embedding_model.py,sha256=0dDL87len4vZ4DR6eCp7JZJCJpgwWphRmJhMK3Se8f4,281
 openai/types/file_content.py,sha256=qLlM4J8kgu1BfrtlmYftPsQVCJu4VqYeiS1T28u8EQ8,184
-openai/types/file_create_params.py,sha256=f8-xfoAlZPl5FuOz0h5sJTdAoBuJEIXVz_iyL9iTCbg,926
+openai/types/file_create_params.py,sha256=N1I3rER1se27usx46fhkvdtn-blJ6Y9ECT7Wwzve37Q,913
 openai/types/file_deleted.py,sha256=H_r9U7XthT5xHAo_4ay1EGGkc21eURt8MkkIBRYiQcw,277
-openai/types/file_list_params.py,sha256=VhZbSrCO0fYnUTgPE_nuBy-3A5MjpXiBtI-BahAc5SY,310
+openai/types/file_list_params.py,sha256=TmmqvM7droAJ49YlgpeFzrhPv5uVkSZDxqlG6hhumPo,960
 openai/types/file_object.py,sha256=ESuRYCTLbDtHxyuhzybKTF_TztIcq_F7TzCTQ6JToE0,1309
+openai/types/file_purpose.py,sha256=o1TzR-41XsNsQ0791GTGPe3DLkU9FEODucKdP6Q6sPc,243
 openai/types/fine_tuning/__init__.py,sha256=SZvjq_22oY9E4zcnrvVd0ul9U4sk_IBeOd0MsNALu5s,806
-openai/types/fine_tuning/fine_tuning_job.py,sha256=YOcsIJZPPAqOnQudOkS_Am-peQuHyyvcMWVDxFvJdEA,3861
-openai/types/fine_tuning/fine_tuning_job_event.py,sha256=oCkO0yImLZnZQLeU4GH6YyUlDG25pzs41SCWWB-sd_o,374
+openai/types/fine_tuning/fine_tuning_job.py,sha256=bu-afb1RZqgNmpUQ7MoXymTjFs3i5JSsBLMV4TKHhi8,6473
+openai/types/fine_tuning/fine_tuning_job_event.py,sha256=POxSD7-WxAtJV2KuEpA9EmZi7W_u0PikOUtUzxIXii4,854
 openai/types/fine_tuning/fine_tuning_job_integration.py,sha256=c3Uy7RMVJ32Xlat-6s9eG-5vZLl4w66COXc0B3pWk4g,242
 openai/types/fine_tuning/fine_tuning_job_wandb_integration.py,sha256=YnBeiz14UuhUSpnD0KBj5V143qLvJbDIMcUVWOCBLXY,1026
 openai/types/fine_tuning/fine_tuning_job_wandb_integration_object.py,sha256=7vEc2uEV2c_DENBjhq0Qy5X8B-rzxsKvGECjnvF1Wdw,804
-openai/types/fine_tuning/job_create_params.py,sha256=Qhclk88QYCmmeBsPzfKBjNHHR5juW6-sUcRZDoSQH94,4693
+openai/types/fine_tuning/job_create_params.py,sha256=TwQlyQrZfxrgqD7nmJDWE8pwklsdUUmkYaitvB7LY34,7222
 openai/types/fine_tuning/job_list_events_params.py,sha256=4xOED4H2ky2mI9sIDytjmfJz5bNAdNWb70WIb_0bBWs,400
 openai/types/fine_tuning/job_list_params.py,sha256=yjxaEnESVTRpJ9ItvjKq30KcD_xz_trqKMIxG2eAriE,396
 openai/types/fine_tuning/jobs/__init__.py,sha256=nuWhOUsmsoVKTKMU35kknmr8sfpTF-kkIzyuOlRbJj0,295
 openai/types/fine_tuning/jobs/checkpoint_list_params.py,sha256=XoDLkkKCWmf5an5rnoVEpNK8mtQHq1fHw9EqmezfrXM,415
 openai/types/fine_tuning/jobs/fine_tuning_job_checkpoint.py,sha256=Z_sUhebJY9nWSssZU7QoOJwe5sez76sCAuVeSO63XhY,1347
 openai/types/image.py,sha256=9No-8GHesOUbjchemY1jqtMwh_s22oBmLVFlLn2KoQo,607
-openai/types/image_create_variation_params.py,sha256=9FuF7N6Ju7BusvbQnMY5ddqHN_YInHkUlqaiVstxwYs,1477
-openai/types/image_edit_params.py,sha256=LvbWaTXVG_yneNnnpkNAocImIhqR-0jaHrRDlj7Tl2I,1837
-openai/types/image_generate_params.py,sha256=S1aA2trSzhLl2OXaFHtQiuJz6P7F_IIzPIswbvUYCjU,2132
+openai/types/image_create_variation_params.py,sha256=PvvPvHXvz0etrRrzVIyvRjvDvNbjGspPu85hOq2fLII,1477
+openai/types/image_edit_params.py,sha256=cxpBybs5peY0DJMTWHgoIx3dWIXj0Y0YmvgxrjGmWjo,1837
+openai/types/image_generate_params.py,sha256=bD2AEIetbt37YDp65vEFfGxkLndOFCwhzJol1I63wfA,2132
 openai/types/image_model.py,sha256=W4YchkhJT2wZdlNDUpVkEKg8zdDDfp9S3oTf4D8Wr8g,219
 openai/types/images_response.py,sha256=EJ4qxYZ8CPGh2SZdRsyw6I0FnUvlgwxwc4NgPovJrvk,274
 openai/types/model.py,sha256=DMw8KwQx8B6S6sAI038D0xdzkmYdY5-r0oMhCUG4l6w,532
 openai/types/model_deleted.py,sha256=tXZybg03DunoOSYvwhT7zKj7KTN42R0VEs_-3PRliMo,229
-openai/types/moderation.py,sha256=ihR2jzld_BfOaHW1_6A2csTInEaJvAl5nPxuh_jegY4,3933
-openai/types/moderation_create_params.py,sha256=TADBGDorBDzcTzkylSB2eoN4cvRmZ0ADN00DzPdI1IA,948
+openai/types/moderation.py,sha256=6CZmxhZiafnT50gKa7BeybrTSoYfCAk7wvD5CQHvBP0,6789
+openai/types/moderation_create_params.py,sha256=EaZ2cej25g5WbRB2kIY7JFCXQPKSQQ95iyoUAAelGr4,992
 openai/types/moderation_create_response.py,sha256=e6SVfWX2_JX25Za0C6KojcnbMTtDB2A7cjUm6cFMKcs,484
-openai/types/moderation_model.py,sha256=zak2cYrNYevj0TItwsa2inX8NhQS0rUJ2Duhsbl7PxU,257
+openai/types/moderation_image_url_input_param.py,sha256=t1r9WD3c-CK2Al1lpB4-DjfzLFSwgETR0g8nsRdoL0Y,622
+openai/types/moderation_model.py,sha256=BFeqSyel2My2WKC6MCa_mAIHJx4uXU3-p8UNudJANeM,319
+openai/types/moderation_multi_modal_input_param.py,sha256=RFdiEPsakWIscutX896ir5_rnEA2TLX5xQkjO5QR2vs,483
+openai/types/moderation_text_input_param.py,sha256=ardCbBcdaULf8bkFuzkSKukV9enrINSjNWvb7m0LjZg,406
 openai/types/shared/__init__.py,sha256=34RJ2IUXj0f3B73a6rqeHILu8AH5-sC8npTbEx_bnk8,551
 openai/types/shared/error_object.py,sha256=G7SGPZ9Qw3gewTKbi3fK69eM6L2Ur0C2D57N8iEapJA,305
 openai/types/shared/function_definition.py,sha256=8a5uHoIKrkrwTgfwTyE9ly4PgsZ3iLA_yRUAjubTb7Y,1447
@@ -608,15 +793,16 @@ openai/types/shared/response_format_json_object.py,sha256=15KTCXJ0o1W4c5V1vAcOQA
 openai/types/shared/response_format_json_schema.py,sha256=rZS7diOPeqK48O_R6OYMJ6AtSGy_88PKTxzha6_56Fo,1399
 openai/types/shared/response_format_text.py,sha256=GX0u_40OLmDdSyawDrUcUk4jcrz1qWsKmmAMP4AD7hc,318
 openai/types/shared_params/__init__.py,sha256=GcNBmK_EPlGE-xPFmSQjlOq7SuNYd2nwDswX4ExHwoU,498
-openai/types/shared_params/function_definition.py,sha256=JM6luT50hXSYChpFmO5aI1XI3E8uhdibYpQrL3NzKvc,1479
+openai/types/shared_params/function_definition.py,sha256=ciMXqn1tFXnp1tg9weJW0uvtyvMLrnph3WXMg4IG1Vk,1482
 openai/types/shared_params/function_parameters.py,sha256=UvxKz_3b9b5ECwWr8RFrIH511htbU2JZsp9Z9BMkF-o,272
 openai/types/shared_params/response_format_json_object.py,sha256=QT4uJCK7RzN3HK17eGjEo36jLKOIBBNGjiX-zIa9iT4,390
 openai/types/shared_params/response_format_json_schema.py,sha256=Uu2ioeSbI64bm-jJ61OY8Lr3PpofTR4d2LNBcaYxlec,1360
 openai/types/shared_params/response_format_text.py,sha256=SjHeZAfgM1-HXAoKLrkiH-VZEnQ73XPTk_RgtJmEbU4,364
 openai/types/upload.py,sha256=mEeQTGS0uqFkxbDpJzgBUvuDhGVPw9cQxhRJjPBVeLo,1186
 openai/types/upload_complete_params.py,sha256=7On-iVAlA9p_nksLSFPBPR4QbB0xEtAW-skyh7S9gR0,504
-openai/types/upload_create_params.py,sha256=RGqqr4UwsQJiB4i_uSm9CcPok8XYPX7c18C95gATS2o,901
+openai/types/upload_create_params.py,sha256=ZiZr1yC6g2VqL7KEnw7lhE4kZvU-F3DfTAc2TPk-XBo,889
 openai/types/uploads/__init__.py,sha256=fDsmd3L0nIWbFldbViOLvcQavsFA4SL3jsXDfAueAck,242
 openai/types/uploads/part_create_params.py,sha256=pBByUzngaj70ov1knoSo_gpeBjaWP9D5EdiHwiG4G7U,362
 openai/types/uploads/upload_part.py,sha256=U9953cr9lJJLWEfhTiwHphRzLKARq3gWAWqrjxbhTR4,590
+openai/types/websocket_connection_options.py,sha256=4cAWpv1KKp_9pvnez7pGYzO3s8zh1WvX2xpBhpe-96k,1840
 openai/version.py,sha256=cjbXKO8Ut3aiv4YlQnugff7AdC48MpSndcx96q88Yb8,62
diff --git a/portkey_ai/_vendor/openai-1.40.1.dist-info/REQUESTED b/portkey_ai/_vendor/openai-1.58.1.dist-info/REQUESTED
similarity index 100%
rename from portkey_ai/_vendor/openai-1.40.1.dist-info/REQUESTED
rename to portkey_ai/_vendor/openai-1.58.1.dist-info/REQUESTED
diff --git a/portkey_ai/_vendor/openai-1.40.1.dist-info/WHEEL b/portkey_ai/_vendor/openai-1.58.1.dist-info/WHEEL
similarity index 67%
rename from portkey_ai/_vendor/openai-1.40.1.dist-info/WHEEL
rename to portkey_ai/_vendor/openai-1.58.1.dist-info/WHEEL
index cdd68a49..12228d41 100644
--- a/portkey_ai/_vendor/openai-1.40.1.dist-info/WHEEL
+++ b/portkey_ai/_vendor/openai-1.58.1.dist-info/WHEEL
@@ -1,4 +1,4 @@
 Wheel-Version: 1.0
-Generator: hatchling 1.25.0
+Generator: hatchling 1.27.0
 Root-Is-Purelib: true
 Tag: py3-none-any
diff --git a/portkey_ai/_vendor/openai-1.40.1.dist-info/entry_points.txt b/portkey_ai/_vendor/openai-1.58.1.dist-info/entry_points.txt
similarity index 100%
rename from portkey_ai/_vendor/openai-1.40.1.dist-info/entry_points.txt
rename to portkey_ai/_vendor/openai-1.58.1.dist-info/entry_points.txt
diff --git a/portkey_ai/_vendor/openai-1.40.1.dist-info/licenses/LICENSE b/portkey_ai/_vendor/openai-1.58.1.dist-info/licenses/LICENSE
similarity index 100%
rename from portkey_ai/_vendor/openai-1.40.1.dist-info/licenses/LICENSE
rename to portkey_ai/_vendor/openai-1.58.1.dist-info/licenses/LICENSE
diff --git a/portkey_ai/_vendor/openai/__init__.py b/portkey_ai/_vendor/openai/__init__.py
index 3c1ebb57..21c60f7e 100644
--- a/portkey_ai/_vendor/openai/__init__.py
+++ b/portkey_ai/_vendor/openai/__init__.py
@@ -6,7 +6,7 @@
 from typing_extensions import override
 
 from . import types
-from ._types import NOT_GIVEN, NoneType, NotGiven, Transport, ProxiesTypes
+from ._types import NOT_GIVEN, Omit, NoneType, NotGiven, Transport, ProxiesTypes
 from ._utils import file_from_path
 from ._client import Client, OpenAI, Stream, Timeout, Transport, AsyncClient, AsyncOpenAI, AsyncStream, RequestOptions
 from ._models import BaseModel
@@ -43,6 +43,7 @@
     "ProxiesTypes",
     "NotGiven",
     "NOT_GIVEN",
+    "Omit",
     "OpenAIError",
     "APIError",
     "APIStatusError",
diff --git a/portkey_ai/_vendor/openai/_base_client.py b/portkey_ai/_vendor/openai/_base_client.py
index 21053e28..cceec903 100644
--- a/portkey_ai/_vendor/openai/_base_client.py
+++ b/portkey_ai/_vendor/openai/_base_client.py
@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import sys
 import json
 import time
 import uuid
@@ -61,7 +62,7 @@
     HttpxRequestFiles,
     ModelBuilderProtocol,
 )
-from ._utils import is_dict, is_list, asyncify, is_given, lru_cache, is_mapping
+from ._utils import SensitiveHeadersFilter, is_dict, is_list, asyncify, is_given, lru_cache, is_mapping
 from ._compat import model_copy, model_dump
 from ._models import GenericModel, FinalRequestOptions, validate_type, construct_type
 from ._response import (
@@ -89,6 +90,7 @@
 from ._legacy_response import LegacyAPIResponse
 
 log: logging.Logger = logging.getLogger(__name__)
+log.addFilter(SensitiveHeadersFilter())
 
 # TODO: make base page type vars covariant
 SyncPageT = TypeVar("SyncPageT", bound="BaseSyncPage[Any]")
@@ -143,6 +145,12 @@ def __init__(
         self.url = url
         self.params = params
 
+    @override
+    def __repr__(self) -> str:
+        if self.url:
+            return f"{self.__class__.__name__}(url={self.url})"
+        return f"{self.__class__.__name__}(params={self.params})"
+
 
 class BasePage(GenericModel, Generic[_T]):
     """
@@ -400,14 +408,7 @@ def _make_status_error(
     ) -> _exceptions.APIStatusError:
         raise NotImplementedError()
 
-    def _remaining_retries(
-        self,
-        remaining_retries: Optional[int],
-        options: FinalRequestOptions,
-    ) -> int:
-        return remaining_retries if remaining_retries is not None else options.get_max_retries(self.max_retries)
-
-    def _build_headers(self, options: FinalRequestOptions) -> httpx.Headers:
+    def _build_headers(self, options: FinalRequestOptions, *, retries_taken: int = 0) -> httpx.Headers:
         custom_headers = options.headers or {}
         headers_dict = _merge_mappings(self.default_headers, custom_headers)
         self._validate_headers(headers_dict, custom_headers)
@@ -419,6 +420,11 @@ def _build_headers(self, options: FinalRequestOptions) -> httpx.Headers:
         if idempotency_header and options.method.lower() != "get" and idempotency_header not in headers:
             headers[idempotency_header] = options.idempotency_key or self._idempotency_key()
 
+        # Don't set the retry count header if it was already set or removed by the caller. We check
+        # `custom_headers`, which can contain `Omit()`, instead of `headers` to account for the removal case.
+        if "x-stainless-retry-count" not in (header.lower() for header in custom_headers):
+            headers["x-stainless-retry-count"] = str(retries_taken)
+
         return headers
 
     def _prepare_url(self, url: str) -> URL:
@@ -440,6 +446,8 @@ def _make_sse_decoder(self) -> SSEDecoder | SSEBytesDecoder:
     def _build_request(
         self,
         options: FinalRequestOptions,
+        *,
+        retries_taken: int = 0,
     ) -> httpx.Request:
         if log.isEnabledFor(logging.DEBUG):
             log.debug("Request options: %s", model_dump(options, exclude_unset=True))
@@ -455,7 +463,7 @@ def _build_request(
             else:
                 raise RuntimeError(f"Unexpected JSON data type, {type(json_data)}, cannot merge with `extra_body`")
 
-        headers = self._build_headers(options)
+        headers = self._build_headers(options, retries_taken=retries_taken)
         params = _merge_mappings(self.default_query, options.params)
         content_type = headers.get("Content-Type")
         files = options.files
@@ -489,12 +497,17 @@ def _build_request(
             if not files:
                 files = cast(HttpxRequestFiles, ForceMultipartDict())
 
+        prepared_url = self._prepare_url(options.url)
+        if "_" in prepared_url.host:
+            # work around https://github.com/encode/httpx/discussions/2880
+            kwargs["extensions"] = {"sni_hostname": prepared_url.host.replace("_", "-")}
+
         # TODO: report this error to httpx
         return self._client.build_request(  # pyright: ignore[reportUnknownMemberType]
             headers=headers,
             timeout=self.timeout if isinstance(options.timeout, NotGiven) else options.timeout,
             method=options.method,
-            url=self._prepare_url(options.url),
+            url=prepared_url,
             # the `Query` type that we use is incompatible with qs'
             # `Params` type as it needs to be typed as `Mapping[str, object]`
             # so that passing a `TypedDict` doesn't cause an error.
@@ -684,7 +697,8 @@ def _calculate_retry_timeout(
         if retry_after is not None and 0 < retry_after <= 60:
             return retry_after
 
-        nb_retries = max_retries - remaining_retries
+        # Also cap retry count to 1000 to avoid any potential overflows with `pow`
+        nb_retries = min(max_retries - remaining_retries, 1000)
 
         # Apply exponential backoff, but not more than the max.
         sleep_seconds = min(INITIAL_RETRY_DELAY * pow(2.0, nb_retries), MAX_RETRY_DELAY)
@@ -849,7 +863,7 @@ def __init__(
             timeout=cast(Timeout, timeout),
             limits=limits,
             follow_redirects=True,
-            **kwargs, # type: ignore
+            **kwargs,  # type: ignore
         )
 
     def is_closed(self) -> bool:
@@ -935,12 +949,17 @@ def request(
         stream: bool = False,
         stream_cls: type[_StreamT] | None = None,
     ) -> ResponseT | _StreamT:
+        if remaining_retries is not None:
+            retries_taken = options.get_max_retries(self.max_retries) - remaining_retries
+        else:
+            retries_taken = 0
+
         return self._request(
             cast_to=cast_to,
             options=options,
             stream=stream,
             stream_cls=stream_cls,
-            remaining_retries=remaining_retries,
+            retries_taken=retries_taken,
         )
 
     def _request(
@@ -948,7 +967,7 @@ def _request(
         *,
         cast_to: Type[ResponseT],
         options: FinalRequestOptions,
-        remaining_retries: int | None,
+        retries_taken: int,
         stream: bool,
         stream_cls: type[_StreamT] | None,
     ) -> ResponseT | _StreamT:
@@ -960,8 +979,8 @@ def _request(
         cast_to = self._maybe_override_cast_to(cast_to, options)
         options = self._prepare_options(options)
 
-        retries = self._remaining_retries(remaining_retries, options)
-        request = self._build_request(options)
+        remaining_retries = options.get_max_retries(self.max_retries) - retries_taken
+        request = self._build_request(options, retries_taken=retries_taken)
         self._prepare_request(request)
 
         kwargs: HttpxSendArgs = {}
@@ -979,11 +998,11 @@ def _request(
         except httpx.TimeoutException as err:
             log.debug("Encountered httpx.TimeoutException", exc_info=True)
 
-            if retries > 0:
+            if remaining_retries > 0:
                 return self._retry_request(
                     input_options,
                     cast_to,
-                    retries,
+                    retries_taken=retries_taken,
                     stream=stream,
                     stream_cls=stream_cls,
                     response_headers=None,
@@ -994,11 +1013,11 @@ def _request(
         except Exception as err:
             log.debug("Encountered Exception", exc_info=True)
 
-            if retries > 0:
+            if remaining_retries > 0:
                 return self._retry_request(
                     input_options,
                     cast_to,
-                    retries,
+                    retries_taken=retries_taken,
                     stream=stream,
                     stream_cls=stream_cls,
                     response_headers=None,
@@ -1022,13 +1041,13 @@ def _request(
         except httpx.HTTPStatusError as err:  # thrown on 4xx and 5xx status code
             log.debug("Encountered httpx.HTTPStatusError", exc_info=True)
 
-            if retries > 0 and self._should_retry(err.response):
+            if remaining_retries > 0 and self._should_retry(err.response):
                 err.response.close()
                 return self._retry_request(
                     input_options,
                     cast_to,
-                    retries,
-                    err.response.headers,
+                    retries_taken=retries_taken,
+                    response_headers=err.response.headers,
                     stream=stream,
                     stream_cls=stream_cls,
                 )
@@ -1047,26 +1066,26 @@ def _request(
             response=response,
             stream=stream,
             stream_cls=stream_cls,
-            retries_taken=options.get_max_retries(self.max_retries) - retries,
+            retries_taken=retries_taken,
         )
 
     def _retry_request(
         self,
         options: FinalRequestOptions,
         cast_to: Type[ResponseT],
-        remaining_retries: int,
-        response_headers: httpx.Headers | None,
         *,
+        retries_taken: int,
+        response_headers: httpx.Headers | None,
         stream: bool,
         stream_cls: type[_StreamT] | None,
     ) -> ResponseT | _StreamT:
-        remaining = remaining_retries - 1
-        if remaining == 1:
+        remaining_retries = options.get_max_retries(self.max_retries) - retries_taken
+        if remaining_retries == 1:
             log.debug("1 retry left")
         else:
-            log.debug("%i retries left", remaining)
+            log.debug("%i retries left", remaining_retries)
 
-        timeout = self._calculate_retry_timeout(remaining, options, response_headers)
+        timeout = self._calculate_retry_timeout(remaining_retries, options, response_headers)
         log.info("Retrying request to %s in %f seconds", options.url, timeout)
 
         # In a synchronous context we are blocking the entire thread. Up to the library user to run the client in a
@@ -1076,7 +1095,7 @@ def _retry_request(
         return self._request(
             options=options,
             cast_to=cast_to,
-            remaining_retries=remaining,
+            retries_taken=retries_taken + 1,
             stream=stream,
             stream_cls=stream_cls,
         )
@@ -1427,7 +1446,7 @@ def __init__(
             timeout=cast(Timeout, timeout),
             limits=limits,
             follow_redirects=True,
-            **kwargs, # type: ignore
+            **kwargs,  # type: ignore
         )
 
     def is_closed(self) -> bool:
@@ -1510,12 +1529,17 @@ async def request(
         stream_cls: type[_AsyncStreamT] | None = None,
         remaining_retries: Optional[int] = None,
     ) -> ResponseT | _AsyncStreamT:
+        if remaining_retries is not None:
+            retries_taken = options.get_max_retries(self.max_retries) - remaining_retries
+        else:
+            retries_taken = 0
+
         return await self._request(
             cast_to=cast_to,
             options=options,
             stream=stream,
             stream_cls=stream_cls,
-            remaining_retries=remaining_retries,
+            retries_taken=retries_taken,
         )
 
     async def _request(
@@ -1525,7 +1549,7 @@ async def _request(
         *,
         stream: bool,
         stream_cls: type[_AsyncStreamT] | None,
-        remaining_retries: int | None,
+        retries_taken: int,
     ) -> ResponseT | _AsyncStreamT:
         if self._platform is None:
             # `get_platform` can make blocking IO calls so we
@@ -1540,8 +1564,8 @@ async def _request(
         cast_to = self._maybe_override_cast_to(cast_to, options)
         options = await self._prepare_options(options)
 
-        retries = self._remaining_retries(remaining_retries, options)
-        request = self._build_request(options)
+        remaining_retries = options.get_max_retries(self.max_retries) - retries_taken
+        request = self._build_request(options, retries_taken=retries_taken)
         await self._prepare_request(request)
 
         kwargs: HttpxSendArgs = {}
@@ -1557,11 +1581,11 @@ async def _request(
         except httpx.TimeoutException as err:
             log.debug("Encountered httpx.TimeoutException", exc_info=True)
 
-            if retries > 0:
+            if remaining_retries > 0:
                 return await self._retry_request(
                     input_options,
                     cast_to,
-                    retries,
+                    retries_taken=retries_taken,
                     stream=stream,
                     stream_cls=stream_cls,
                     response_headers=None,
@@ -1572,11 +1596,11 @@ async def _request(
         except Exception as err:
             log.debug("Encountered Exception", exc_info=True)
 
-            if retries > 0:
+            if remaining_retries > 0:
                 return await self._retry_request(
                     input_options,
                     cast_to,
-                    retries,
+                    retries_taken=retries_taken,
                     stream=stream,
                     stream_cls=stream_cls,
                     response_headers=None,
@@ -1594,13 +1618,13 @@ async def _request(
         except httpx.HTTPStatusError as err:  # thrown on 4xx and 5xx status code
             log.debug("Encountered httpx.HTTPStatusError", exc_info=True)
 
-            if retries > 0 and self._should_retry(err.response):
+            if remaining_retries > 0 and self._should_retry(err.response):
                 await err.response.aclose()
                 return await self._retry_request(
                     input_options,
                     cast_to,
-                    retries,
-                    err.response.headers,
+                    retries_taken=retries_taken,
+                    response_headers=err.response.headers,
                     stream=stream,
                     stream_cls=stream_cls,
                 )
@@ -1619,26 +1643,26 @@ async def _request(
             response=response,
             stream=stream,
             stream_cls=stream_cls,
-            retries_taken=options.get_max_retries(self.max_retries) - retries,
+            retries_taken=retries_taken,
         )
 
     async def _retry_request(
         self,
         options: FinalRequestOptions,
         cast_to: Type[ResponseT],
-        remaining_retries: int,
-        response_headers: httpx.Headers | None,
         *,
+        retries_taken: int,
+        response_headers: httpx.Headers | None,
         stream: bool,
         stream_cls: type[_AsyncStreamT] | None,
     ) -> ResponseT | _AsyncStreamT:
-        remaining = remaining_retries - 1
-        if remaining == 1:
+        remaining_retries = options.get_max_retries(self.max_retries) - retries_taken
+        if remaining_retries == 1:
             log.debug("1 retry left")
         else:
-            log.debug("%i retries left", remaining)
+            log.debug("%i retries left", remaining_retries)
 
-        timeout = self._calculate_retry_timeout(remaining, options, response_headers)
+        timeout = self._calculate_retry_timeout(remaining_retries, options, response_headers)
         log.info("Retrying request to %s in %f seconds", options.url, timeout)
 
         await anyio.sleep(timeout)
@@ -1646,7 +1670,7 @@ async def _retry_request(
         return await self._request(
             options=options,
             cast_to=cast_to,
-            remaining_retries=remaining,
+            retries_taken=retries_taken + 1,
             stream=stream,
             stream_cls=stream_cls,
         )
@@ -2016,7 +2040,6 @@ def get_python_version() -> str:
 
 def get_architecture() -> Arch:
     try:
-        python_bitness, _ = platform.architecture()
         machine = platform.machine().lower()
     except Exception:
         return "unknown"
@@ -2032,7 +2055,7 @@ def get_architecture() -> Arch:
         return "x64"
 
     # TODO: untested
-    if python_bitness == "32bit":
+    if sys.maxsize <= 2**32:
         return "x32"
 
     if machine:
diff --git a/portkey_ai/_vendor/openai/_client.py b/portkey_ai/_vendor/openai/_client.py
index a1f7dde4..c784694f 100644
--- a/portkey_ai/_vendor/openai/_client.py
+++ b/portkey_ai/_vendor/openai/_client.py
@@ -8,7 +8,7 @@
 
 import httpx
 
-from . import resources, _exceptions
+from . import _exceptions
 from ._qs import Querystring
 from ._types import (
     NOT_GIVEN,
@@ -25,6 +25,7 @@
     get_async_library,
 )
 from ._version import __version__
+from .resources import files, images, models, batches, embeddings, completions, moderations
 from ._streaming import Stream as Stream, AsyncStream as AsyncStream
 from ._exceptions import OpenAIError, APIStatusError
 from ._base_client import (
@@ -32,33 +33,28 @@
     SyncAPIClient,
     AsyncAPIClient,
 )
+from .resources.beta import beta
+from .resources.chat import chat
+from .resources.audio import audio
+from .resources.uploads import uploads
+from .resources.fine_tuning import fine_tuning
 
-__all__ = [
-    "Timeout",
-    "Transport",
-    "ProxiesTypes",
-    "RequestOptions",
-    "resources",
-    "OpenAI",
-    "AsyncOpenAI",
-    "Client",
-    "AsyncClient",
-]
+__all__ = ["Timeout", "Transport", "ProxiesTypes", "RequestOptions", "OpenAI", "AsyncOpenAI", "Client", "AsyncClient"]
 
 
 class OpenAI(SyncAPIClient):
-    completions: resources.Completions
-    chat: resources.Chat
-    embeddings: resources.Embeddings
-    files: resources.Files
-    images: resources.Images
-    audio: resources.Audio
-    moderations: resources.Moderations
-    models: resources.Models
-    fine_tuning: resources.FineTuning
-    beta: resources.Beta
-    batches: resources.Batches
-    uploads: resources.Uploads # TODO
+    completions: completions.Completions
+    chat: chat.Chat
+    embeddings: embeddings.Embeddings
+    files: files.Files
+    images: images.Images
+    audio: audio.Audio
+    moderations: moderations.Moderations
+    models: models.Models
+    fine_tuning: fine_tuning.FineTuning
+    beta: beta.Beta
+    batches: batches.Batches
+    uploads: uploads.Uploads
     with_raw_response: OpenAIWithRawResponse
     with_streaming_response: OpenAIWithStreamedResponse
 
@@ -67,6 +63,14 @@ class OpenAI(SyncAPIClient):
     organization: str | None
     project: str | None
 
+    websocket_base_url: str | httpx.URL | None
+    """Base URL for WebSocket connections.
+
+    If not specified, the default base URL will be used, with 'wss://' replacing the
+    'http://' or 'https://' scheme. For example: 'http://example.com' becomes
+    'wss://example.com'
+    """
+
     def __init__(
         self,
         *,
@@ -74,6 +78,7 @@ def __init__(
         organization: str | None = None,
         project: str | None = None,
         base_url: str | httpx.URL | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         timeout: Union[float, Timeout, None, NotGiven] = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
         default_headers: Mapping[str, str] | None = None,
@@ -115,6 +120,8 @@ def __init__(
             project = os.environ.get("OPENAI_PROJECT_ID")
         self.project = project
 
+        self.websocket_base_url = websocket_base_url
+
         if base_url is None:
             base_url = os.environ.get("OPENAI_BASE_URL")
         if base_url is None:
@@ -133,18 +140,18 @@ def __init__(
 
         self._default_stream_cls = Stream
 
-        self.completions = resources.Completions(self)
-        self.chat = resources.Chat(self)
-        self.embeddings = resources.Embeddings(self)
-        self.files = resources.Files(self)
-        self.images = resources.Images(self)
-        self.audio = resources.Audio(self)
-        self.moderations = resources.Moderations(self)
-        self.models = resources.Models(self)
-        self.fine_tuning = resources.FineTuning(self)
-        self.beta = resources.Beta(self)
-        self.batches = resources.Batches(self)
-        self.uploads = resources.Uploads(self)
+        self.completions = completions.Completions(self)
+        self.chat = chat.Chat(self)
+        self.embeddings = embeddings.Embeddings(self)
+        self.files = files.Files(self)
+        self.images = images.Images(self)
+        self.audio = audio.Audio(self)
+        self.moderations = moderations.Moderations(self)
+        self.models = models.Models(self)
+        self.fine_tuning = fine_tuning.FineTuning(self)
+        self.beta = beta.Beta(self)
+        self.batches = batches.Batches(self)
+        self.uploads = uploads.Uploads(self)
         self.with_raw_response = OpenAIWithRawResponse(self)
         self.with_streaming_response = OpenAIWithStreamedResponse(self)
 
@@ -176,6 +183,7 @@ def copy(
         api_key: str | None = None,
         organization: str | None = None,
         project: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         base_url: str | httpx.URL | None = None,
         timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
         http_client: httpx.Client | None = None,
@@ -212,6 +220,7 @@ def copy(
             api_key=api_key or self.api_key,
             organization=organization or self.organization,
             project=project or self.project,
+            websocket_base_url=websocket_base_url or self.websocket_base_url,
             base_url=base_url or self.base_url,
             timeout=self.timeout if isinstance(timeout, NotGiven) else timeout,
             http_client=http_client,
@@ -261,18 +270,18 @@ def _make_status_error(
 
 
 class AsyncOpenAI(AsyncAPIClient):
-    completions: resources.AsyncCompletions
-    chat: resources.AsyncChat
-    embeddings: resources.AsyncEmbeddings
-    files: resources.AsyncFiles
-    images: resources.AsyncImages
-    audio: resources.AsyncAudio
-    moderations: resources.AsyncModerations
-    models: resources.AsyncModels
-    fine_tuning: resources.AsyncFineTuning
-    beta: resources.AsyncBeta
-    batches: resources.AsyncBatches
-    uploads: resources.AsyncUploads
+    completions: completions.AsyncCompletions
+    chat: chat.AsyncChat
+    embeddings: embeddings.AsyncEmbeddings
+    files: files.AsyncFiles
+    images: images.AsyncImages
+    audio: audio.AsyncAudio
+    moderations: moderations.AsyncModerations
+    models: models.AsyncModels
+    fine_tuning: fine_tuning.AsyncFineTuning
+    beta: beta.AsyncBeta
+    batches: batches.AsyncBatches
+    uploads: uploads.AsyncUploads
     with_raw_response: AsyncOpenAIWithRawResponse
     with_streaming_response: AsyncOpenAIWithStreamedResponse
 
@@ -281,6 +290,14 @@ class AsyncOpenAI(AsyncAPIClient):
     organization: str | None
     project: str | None
 
+    websocket_base_url: str | httpx.URL | None
+    """Base URL for WebSocket connections.
+
+    If not specified, the default base URL will be used, with 'wss://' replacing the
+    'http://' or 'https://' scheme. For example: 'http://example.com' becomes
+    'wss://example.com'
+    """
+
     def __init__(
         self,
         *,
@@ -288,6 +305,7 @@ def __init__(
         organization: str | None = None,
         project: str | None = None,
         base_url: str | httpx.URL | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         timeout: Union[float, Timeout, None, NotGiven] = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
         default_headers: Mapping[str, str] | None = None,
@@ -329,6 +347,8 @@ def __init__(
             project = os.environ.get("OPENAI_PROJECT_ID")
         self.project = project
 
+        self.websocket_base_url = websocket_base_url
+
         if base_url is None:
             base_url = os.environ.get("OPENAI_BASE_URL")
         if base_url is None:
@@ -347,18 +367,18 @@ def __init__(
 
         self._default_stream_cls = AsyncStream
 
-        self.completions = resources.AsyncCompletions(self)
-        self.chat = resources.AsyncChat(self)
-        self.embeddings = resources.AsyncEmbeddings(self)
-        self.files = resources.AsyncFiles(self)
-        self.images = resources.AsyncImages(self)
-        self.audio = resources.AsyncAudio(self)
-        self.moderations = resources.AsyncModerations(self)
-        self.models = resources.AsyncModels(self)
-        self.fine_tuning = resources.AsyncFineTuning(self)
-        self.beta = resources.AsyncBeta(self)
-        self.batches = resources.AsyncBatches(self)
-        self.uploads = resources.AsyncUploads(self)
+        self.completions = completions.AsyncCompletions(self)
+        self.chat = chat.AsyncChat(self)
+        self.embeddings = embeddings.AsyncEmbeddings(self)
+        self.files = files.AsyncFiles(self)
+        self.images = images.AsyncImages(self)
+        self.audio = audio.AsyncAudio(self)
+        self.moderations = moderations.AsyncModerations(self)
+        self.models = models.AsyncModels(self)
+        self.fine_tuning = fine_tuning.AsyncFineTuning(self)
+        self.beta = beta.AsyncBeta(self)
+        self.batches = batches.AsyncBatches(self)
+        self.uploads = uploads.AsyncUploads(self)
         self.with_raw_response = AsyncOpenAIWithRawResponse(self)
         self.with_streaming_response = AsyncOpenAIWithStreamedResponse(self)
 
@@ -390,6 +410,7 @@ def copy(
         api_key: str | None = None,
         organization: str | None = None,
         project: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         base_url: str | httpx.URL | None = None,
         timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
         http_client: httpx.AsyncClient | None = None,
@@ -426,6 +447,7 @@ def copy(
             api_key=api_key or self.api_key,
             organization=organization or self.organization,
             project=project or self.project,
+            websocket_base_url=websocket_base_url or self.websocket_base_url,
             base_url=base_url or self.base_url,
             timeout=self.timeout if isinstance(timeout, NotGiven) else timeout,
             http_client=http_client,
@@ -476,66 +498,66 @@ def _make_status_error(
 
 class OpenAIWithRawResponse:
     def __init__(self, client: OpenAI) -> None:
-        self.completions = resources.CompletionsWithRawResponse(client.completions)
-        self.chat = resources.ChatWithRawResponse(client.chat)
-        self.embeddings = resources.EmbeddingsWithRawResponse(client.embeddings)
-        self.files = resources.FilesWithRawResponse(client.files)
-        self.images = resources.ImagesWithRawResponse(client.images)
-        self.audio = resources.AudioWithRawResponse(client.audio)
-        self.moderations = resources.ModerationsWithRawResponse(client.moderations)
-        self.models = resources.ModelsWithRawResponse(client.models)
-        self.fine_tuning = resources.FineTuningWithRawResponse(client.fine_tuning)
-        self.beta = resources.BetaWithRawResponse(client.beta)
-        self.batches = resources.BatchesWithRawResponse(client.batches)
-        self.uploads = resources.UploadsWithRawResponse(client.uploads)
+        self.completions = completions.CompletionsWithRawResponse(client.completions)
+        self.chat = chat.ChatWithRawResponse(client.chat)
+        self.embeddings = embeddings.EmbeddingsWithRawResponse(client.embeddings)
+        self.files = files.FilesWithRawResponse(client.files)
+        self.images = images.ImagesWithRawResponse(client.images)
+        self.audio = audio.AudioWithRawResponse(client.audio)
+        self.moderations = moderations.ModerationsWithRawResponse(client.moderations)
+        self.models = models.ModelsWithRawResponse(client.models)
+        self.fine_tuning = fine_tuning.FineTuningWithRawResponse(client.fine_tuning)
+        self.beta = beta.BetaWithRawResponse(client.beta)
+        self.batches = batches.BatchesWithRawResponse(client.batches)
+        self.uploads = uploads.UploadsWithRawResponse(client.uploads)
 
 
 class AsyncOpenAIWithRawResponse:
     def __init__(self, client: AsyncOpenAI) -> None:
-        self.completions = resources.AsyncCompletionsWithRawResponse(client.completions)
-        self.chat = resources.AsyncChatWithRawResponse(client.chat)
-        self.embeddings = resources.AsyncEmbeddingsWithRawResponse(client.embeddings)
-        self.files = resources.AsyncFilesWithRawResponse(client.files)
-        self.images = resources.AsyncImagesWithRawResponse(client.images)
-        self.audio = resources.AsyncAudioWithRawResponse(client.audio)
-        self.moderations = resources.AsyncModerationsWithRawResponse(client.moderations)
-        self.models = resources.AsyncModelsWithRawResponse(client.models)
-        self.fine_tuning = resources.AsyncFineTuningWithRawResponse(client.fine_tuning)
-        self.beta = resources.AsyncBetaWithRawResponse(client.beta)
-        self.batches = resources.AsyncBatchesWithRawResponse(client.batches)
-        self.uploads = resources.AsyncUploadsWithRawResponse(client.uploads)
+        self.completions = completions.AsyncCompletionsWithRawResponse(client.completions)
+        self.chat = chat.AsyncChatWithRawResponse(client.chat)
+        self.embeddings = embeddings.AsyncEmbeddingsWithRawResponse(client.embeddings)
+        self.files = files.AsyncFilesWithRawResponse(client.files)
+        self.images = images.AsyncImagesWithRawResponse(client.images)
+        self.audio = audio.AsyncAudioWithRawResponse(client.audio)
+        self.moderations = moderations.AsyncModerationsWithRawResponse(client.moderations)
+        self.models = models.AsyncModelsWithRawResponse(client.models)
+        self.fine_tuning = fine_tuning.AsyncFineTuningWithRawResponse(client.fine_tuning)
+        self.beta = beta.AsyncBetaWithRawResponse(client.beta)
+        self.batches = batches.AsyncBatchesWithRawResponse(client.batches)
+        self.uploads = uploads.AsyncUploadsWithRawResponse(client.uploads)
 
 
 class OpenAIWithStreamedResponse:
     def __init__(self, client: OpenAI) -> None:
-        self.completions = resources.CompletionsWithStreamingResponse(client.completions)
-        self.chat = resources.ChatWithStreamingResponse(client.chat)
-        self.embeddings = resources.EmbeddingsWithStreamingResponse(client.embeddings)
-        self.files = resources.FilesWithStreamingResponse(client.files)
-        self.images = resources.ImagesWithStreamingResponse(client.images)
-        self.audio = resources.AudioWithStreamingResponse(client.audio)
-        self.moderations = resources.ModerationsWithStreamingResponse(client.moderations)
-        self.models = resources.ModelsWithStreamingResponse(client.models)
-        self.fine_tuning = resources.FineTuningWithStreamingResponse(client.fine_tuning)
-        self.beta = resources.BetaWithStreamingResponse(client.beta)
-        self.batches = resources.BatchesWithStreamingResponse(client.batches)
-        self.uploads = resources.UploadsWithStreamingResponse(client.uploads)
+        self.completions = completions.CompletionsWithStreamingResponse(client.completions)
+        self.chat = chat.ChatWithStreamingResponse(client.chat)
+        self.embeddings = embeddings.EmbeddingsWithStreamingResponse(client.embeddings)
+        self.files = files.FilesWithStreamingResponse(client.files)
+        self.images = images.ImagesWithStreamingResponse(client.images)
+        self.audio = audio.AudioWithStreamingResponse(client.audio)
+        self.moderations = moderations.ModerationsWithStreamingResponse(client.moderations)
+        self.models = models.ModelsWithStreamingResponse(client.models)
+        self.fine_tuning = fine_tuning.FineTuningWithStreamingResponse(client.fine_tuning)
+        self.beta = beta.BetaWithStreamingResponse(client.beta)
+        self.batches = batches.BatchesWithStreamingResponse(client.batches)
+        self.uploads = uploads.UploadsWithStreamingResponse(client.uploads)
 
 
 class AsyncOpenAIWithStreamedResponse:
     def __init__(self, client: AsyncOpenAI) -> None:
-        self.completions = resources.AsyncCompletionsWithStreamingResponse(client.completions)
-        self.chat = resources.AsyncChatWithStreamingResponse(client.chat)
-        self.embeddings = resources.AsyncEmbeddingsWithStreamingResponse(client.embeddings)
-        self.files = resources.AsyncFilesWithStreamingResponse(client.files)
-        self.images = resources.AsyncImagesWithStreamingResponse(client.images)
-        self.audio = resources.AsyncAudioWithStreamingResponse(client.audio)
-        self.moderations = resources.AsyncModerationsWithStreamingResponse(client.moderations)
-        self.models = resources.AsyncModelsWithStreamingResponse(client.models)
-        self.fine_tuning = resources.AsyncFineTuningWithStreamingResponse(client.fine_tuning)
-        self.beta = resources.AsyncBetaWithStreamingResponse(client.beta)
-        self.batches = resources.AsyncBatchesWithStreamingResponse(client.batches)
-        self.uploads = resources.AsyncUploadsWithStreamingResponse(client.uploads)
+        self.completions = completions.AsyncCompletionsWithStreamingResponse(client.completions)
+        self.chat = chat.AsyncChatWithStreamingResponse(client.chat)
+        self.embeddings = embeddings.AsyncEmbeddingsWithStreamingResponse(client.embeddings)
+        self.files = files.AsyncFilesWithStreamingResponse(client.files)
+        self.images = images.AsyncImagesWithStreamingResponse(client.images)
+        self.audio = audio.AsyncAudioWithStreamingResponse(client.audio)
+        self.moderations = moderations.AsyncModerationsWithStreamingResponse(client.moderations)
+        self.models = models.AsyncModelsWithStreamingResponse(client.models)
+        self.fine_tuning = fine_tuning.AsyncFineTuningWithStreamingResponse(client.fine_tuning)
+        self.beta = beta.AsyncBetaWithStreamingResponse(client.beta)
+        self.batches = batches.AsyncBatchesWithStreamingResponse(client.batches)
+        self.uploads = uploads.AsyncUploadsWithStreamingResponse(client.uploads)
 
 
 Client = OpenAI
diff --git a/portkey_ai/_vendor/openai/_compat.py b/portkey_ai/_vendor/openai/_compat.py
index c0dd8c1e..87fc3707 100644
--- a/portkey_ai/_vendor/openai/_compat.py
+++ b/portkey_ai/_vendor/openai/_compat.py
@@ -2,7 +2,7 @@
 
 from typing import TYPE_CHECKING, Any, Union, Generic, TypeVar, Callable, cast, overload
 from datetime import date, datetime
-from typing_extensions import Self
+from typing_extensions import Self, Literal
 
 import pydantic
 from pydantic.fields import FieldInfo
@@ -133,15 +133,20 @@ def model_json(model: pydantic.BaseModel, *, indent: int | None = None) -> str:
 def model_dump(
     model: pydantic.BaseModel,
     *,
-    exclude: IncEx = None,
+    exclude: IncEx | None = None,
     exclude_unset: bool = False,
     exclude_defaults: bool = False,
+    warnings: bool = True,
+    mode: Literal["json", "python"] = "python",
 ) -> dict[str, Any]:
-    if PYDANTIC_V2:
+    if PYDANTIC_V2 or hasattr(model, "model_dump"):
         return model.model_dump(
+            mode=mode,
             exclude=exclude,
             exclude_unset=exclude_unset,
             exclude_defaults=exclude_defaults,
+            # warnings are not supported in Pydantic v1
+            warnings=warnings if PYDANTIC_V2 else True,
         )
     return cast(
         "dict[str, Any]",
@@ -221,9 +226,6 @@ def __set_name__(self, owner: type[Any], name: str) -> None: ...
         # __set__ is not defined at runtime, but @cached_property is designed to be settable
         def __set__(self, instance: object, value: _T) -> None: ...
 else:
-    try:
-        from functools import cached_property as cached_property
-    except ImportError:
-        from cached_property import cached_property as cached_property
+    from functools import cached_property as cached_property
 
     typed_cached_property = cached_property
diff --git a/portkey_ai/_vendor/openai/_exceptions.py b/portkey_ai/_vendor/openai/_exceptions.py
index 3a87c92b..e326ed95 100644
--- a/portkey_ai/_vendor/openai/_exceptions.py
+++ b/portkey_ai/_vendor/openai/_exceptions.py
@@ -2,7 +2,7 @@
 
 from __future__ import annotations
 
-from typing import Any, Optional, cast
+from typing import TYPE_CHECKING, Any, Optional, cast
 from typing_extensions import Literal
 
 import httpx
@@ -10,6 +10,9 @@
 from ._utils import is_dict
 from ._models import construct_type
 
+if TYPE_CHECKING:
+    from .types.chat import ChatCompletion
+
 __all__ = [
     "BadRequestError",
     "AuthenticationError",
@@ -88,11 +91,7 @@ def __init__(self, message: str, *, response: httpx.Response, body: object | Non
 
 
 class APIConnectionError(APIError):
-    LOCALHOST_CONNECTION_ERROR = """Could not instantiate the Portkey client. \
-You can either add a valid `api_key` parameter (from https://app.portkey.ai/api-keys) \
-or check the `base_url` parameter in the Portkey client, for your AI Gateway's instance's URL.
-"""
-    def __init__(self, *, message: str = LOCALHOST_CONNECTION_ERROR, request: httpx.Request) -> None:
+    def __init__(self, *, message: str = "Connection error.", request: httpx.Request) -> None:
         super().__init__(message, request, body=None)
 
 
@@ -134,10 +133,20 @@ class InternalServerError(APIStatusError):
 
 
 class LengthFinishReasonError(OpenAIError):
-    def __init__(self) -> None:
-        super().__init__(
-            f"Could not parse response content as the length limit was reached",
-        )
+    completion: ChatCompletion
+    """The completion that caused this error.
+
+    Note: this will *not* be a complete `ChatCompletion` object when streaming as `usage`
+          will not be included.
+    """
+
+    def __init__(self, *, completion: ChatCompletion) -> None:
+        msg = "Could not parse response content as the length limit was reached"
+        if completion.usage:
+            msg += f" - {completion.usage}"
+
+        super().__init__(msg)
+        self.completion = completion
 
 
 class ContentFilterFinishReasonError(OpenAIError):
diff --git a/portkey_ai/_vendor/openai/_legacy_response.py b/portkey_ai/_vendor/openai/_legacy_response.py
index c42fb8b8..7a14f27a 100644
--- a/portkey_ai/_vendor/openai/_legacy_response.py
+++ b/portkey_ai/_vendor/openai/_legacy_response.py
@@ -24,8 +24,8 @@
 import pydantic
 
 from ._types import NoneType
-from ._utils import is_given, extract_type_arg, is_annotated_type
-from ._models import BaseModel, is_basemodel
+from ._utils import is_given, extract_type_arg, is_annotated_type, is_type_alias_type
+from ._models import BaseModel, is_basemodel, add_request_id
 from ._constants import RAW_RESPONSE_HEADER
 from ._streaming import Stream, AsyncStream, is_stream_class_type, extract_stream_chunk_type
 from ._exceptions import APIResponseValidationError
@@ -138,8 +138,11 @@ class MyModel(BaseModel):
         if is_given(self._options.post_parser):
             parsed = self._options.post_parser(parsed)
 
+        if isinstance(parsed, BaseModel):
+            add_request_id(parsed, self.request_id)
+
         self._parsed_by_type[cache_key] = parsed
-        return parsed
+        return cast(R, parsed)
 
     @property
     def headers(self) -> httpx.Headers:
@@ -192,9 +195,15 @@ def elapsed(self) -> datetime.timedelta:
         return self.http_response.elapsed
 
     def _parse(self, *, to: type[_T] | None = None) -> R | _T:
+        cast_to = to if to is not None else self._cast_to
+
+        # unwrap `TypeAlias('Name', T)` -> `T`
+        if is_type_alias_type(cast_to):
+            cast_to = cast_to.__value__  # type: ignore[unreachable]
+
         # unwrap `Annotated[T, ...]` -> `T`
-        if to and is_annotated_type(to):
-            to = extract_type_arg(to, 0)
+        if cast_to and is_annotated_type(cast_to):
+            cast_to = extract_type_arg(cast_to, 0)
 
         if self._stream:
             if to:
@@ -230,18 +239,12 @@ def _parse(self, *, to: type[_T] | None = None) -> R | _T:
             return cast(
                 R,
                 stream_cls(
-                    cast_to=self._cast_to,
+                    cast_to=cast_to,
                     response=self.http_response,
                     client=cast(Any, self._client),
                 ),
             )
 
-        cast_to = to if to is not None else self._cast_to
-
-        # unwrap `Annotated[T, ...]` -> `T`
-        if is_annotated_type(cast_to):
-            cast_to = extract_type_arg(cast_to, 0)
-
         if cast_to is NoneType:
             return cast(R, None)
 
@@ -255,6 +258,9 @@ def _parse(self, *, to: type[_T] | None = None) -> R | _T:
         if cast_to == float:
             return cast(R, float(response.text))
 
+        if cast_to == bool:
+            return cast(R, response.text.lower() == "true")
+
         origin = get_origin(cast_to) or cast_to
 
         if inspect.isclass(origin) and issubclass(origin, HttpxBinaryResponseContent):
diff --git a/portkey_ai/_vendor/openai/_models.py b/portkey_ai/_vendor/openai/_models.py
index 5148d5a7..2f67e5eb 100644
--- a/portkey_ai/_vendor/openai/_models.py
+++ b/portkey_ai/_vendor/openai/_models.py
@@ -2,7 +2,7 @@
 
 import os
 import inspect
-from typing import TYPE_CHECKING, Any, Type, Union, Generic, TypeVar, Callable, cast
+from typing import TYPE_CHECKING, Any, Type, Tuple, Union, Generic, TypeVar, Callable, Optional, cast
 from datetime import date, datetime
 from typing_extensions import (
     Unpack,
@@ -10,6 +10,7 @@
     ClassVar,
     Protocol,
     Required,
+    Sequence,
     ParamSpec,
     TypedDict,
     TypeGuard,
@@ -37,6 +38,7 @@
     PropertyInfo,
     is_list,
     is_given,
+    json_safe,
     lru_cache,
     is_mapping,
     parse_date,
@@ -45,6 +47,7 @@
     strip_not_given,
     extract_type_arg,
     is_annotated_type,
+    is_type_alias_type,
     strip_annotated_type,
 )
 from ._compat import (
@@ -72,6 +75,8 @@
 
 P = ParamSpec("P")
 
+ReprArgs = Sequence[Tuple[Optional[str], Any]]
+
 
 @runtime_checkable
 class _ConfigProtocol(Protocol):
@@ -94,6 +99,28 @@ def model_fields_set(self) -> set[str]:
         class Config(pydantic.BaseConfig):  # pyright: ignore[reportDeprecated]
             extra: Any = pydantic.Extra.allow  # type: ignore
 
+        @override
+        def __repr_args__(self) -> ReprArgs:
+            # we don't want these attributes to be included when something like `rich.print` is used
+            return [arg for arg in super().__repr_args__() if arg[0] not in {"_request_id", "__exclude_fields__"}]
+
+    if TYPE_CHECKING:
+        _request_id: Optional[str] = None
+        """The ID of the request, returned via the X-Request-ID header. Useful for debugging requests and reporting issues to OpenAI.
+
+        This will **only** be set for the top-level response object, it will not be defined for nested objects. For example:
+        
+        ```py
+        completion = await client.chat.completions.create(...)
+        completion._request_id  # req_id_xxx
+        completion.usage._request_id  # raises `AttributeError`
+        ```
+
+        Note: unlike other properties that use an `_` prefix, this property
+        *is* public. Unless documented otherwise, all other `_` prefix properties,
+        methods and modules are *private*.
+        """
+
     def to_dict(
         self,
         *,
@@ -176,7 +203,7 @@ def __str__(self) -> str:
     # Based on https://github.com/samuelcolvin/pydantic/issues/1168#issuecomment-817742836.
     @classmethod
     @override
-    def construct(
+    def construct(  # pyright: ignore[reportIncompatibleMethodOverride]
         cls: Type[ModelT],
         _fields_set: set[str] | None = None,
         **values: object,
@@ -248,8 +275,8 @@ def model_dump(
             self,
             *,
             mode: Literal["json", "python"] | str = "python",
-            include: IncEx = None,
-            exclude: IncEx = None,
+            include: IncEx | None = None,
+            exclude: IncEx | None = None,
             by_alias: bool = False,
             exclude_unset: bool = False,
             exclude_defaults: bool = False,
@@ -279,8 +306,8 @@ def model_dump(
             Returns:
                 A dictionary representation of the model.
             """
-            if mode != "python":
-                raise ValueError("mode is only supported in Pydantic v2")
+            if mode not in {"json", "python"}:
+                raise ValueError("mode must be either 'json' or 'python'")
             if round_trip != False:
                 raise ValueError("round_trip is only supported in Pydantic v2")
             if warnings != True:
@@ -289,7 +316,7 @@ def model_dump(
                 raise ValueError("context is only supported in Pydantic v2")
             if serialize_as_any != False:
                 raise ValueError("serialize_as_any is only supported in Pydantic v2")
-            return super().dict(  # pyright: ignore[reportDeprecated]
+            dumped = super().dict(  # pyright: ignore[reportDeprecated]
                 include=include,
                 exclude=exclude,
                 by_alias=by_alias,
@@ -298,13 +325,15 @@ def model_dump(
                 exclude_none=exclude_none,
             )
 
+            return cast(dict[str, Any], json_safe(dumped)) if mode == "json" else dumped
+
         @override
         def model_dump_json(
             self,
             *,
             indent: int | None = None,
-            include: IncEx = None,
-            exclude: IncEx = None,
+            include: IncEx | None = None,
+            exclude: IncEx | None = None,
             by_alias: bool = False,
             exclude_unset: bool = False,
             exclude_defaults: bool = False,
@@ -380,6 +409,8 @@ def is_basemodel(type_: type) -> bool:
 
 def is_basemodel_type(type_: type) -> TypeGuard[type[BaseModel] | type[GenericModel]]:
     origin = get_origin(type_) or type_
+    if not inspect.isclass(origin):
+        return False
     return issubclass(origin, BaseModel) or issubclass(origin, GenericModel)
 
 
@@ -423,6 +454,8 @@ def construct_type(*, value: object, type_: object) -> object:
     # we allow `object` as the input type because otherwise, passing things like
     # `Literal['value']` will be reported as a type error by type checkers
     type_ = cast("type[object]", type_)
+    if is_type_alias_type(type_):
+        type_ = type_.__value__  # type: ignore[unreachable]
 
     # unwrap `Annotated[T, ...]` -> `T`
     if is_annotated_type(type_):
@@ -660,6 +693,21 @@ def set_pydantic_config(typ: Any, config: pydantic.ConfigDict) -> None:
     setattr(typ, "__pydantic_config__", config)  # noqa: B010
 
 
+def add_request_id(obj: BaseModel, request_id: str | None) -> None:
+    obj._request_id = request_id
+
+    # in Pydantic v1, using setattr like we do above causes the attribute
+    # to be included when serializing the model which we don't want in this
+    # case so we need to explicitly exclude it
+    if not PYDANTIC_V2:
+        try:
+            exclude_fields = obj.__exclude_fields__  # type: ignore
+        except AttributeError:
+            cast(Any, obj).__exclude_fields__ = {"_request_id", "__exclude_fields__"}
+        else:
+            cast(Any, obj).__exclude_fields__ = {*(exclude_fields or {}), "_request_id", "__exclude_fields__"}
+
+
 # our use of subclasssing here causes weirdness for type checkers,
 # so we just pretend that we don't subclass
 if TYPE_CHECKING:
diff --git a/portkey_ai/_vendor/openai/_response.py b/portkey_ai/_vendor/openai/_response.py
index f9d91786..15274465 100644
--- a/portkey_ai/_vendor/openai/_response.py
+++ b/portkey_ai/_vendor/openai/_response.py
@@ -25,8 +25,8 @@
 import pydantic
 
 from ._types import NoneType
-from ._utils import is_given, extract_type_arg, is_annotated_type, extract_type_var_from_base
-from ._models import BaseModel, is_basemodel
+from ._utils import is_given, extract_type_arg, is_annotated_type, is_type_alias_type, extract_type_var_from_base
+from ._models import BaseModel, is_basemodel, add_request_id
 from ._constants import RAW_RESPONSE_HEADER, OVERRIDE_CAST_TO_HEADER
 from ._streaming import Stream, AsyncStream, is_stream_class_type, extract_stream_chunk_type
 from ._exceptions import OpenAIError, APIResponseValidationError
@@ -126,9 +126,15 @@ def __repr__(self) -> str:
         )
 
     def _parse(self, *, to: type[_T] | None = None) -> R | _T:
+        cast_to = to if to is not None else self._cast_to
+
+        # unwrap `TypeAlias('Name', T)` -> `T`
+        if is_type_alias_type(cast_to):
+            cast_to = cast_to.__value__  # type: ignore[unreachable]
+
         # unwrap `Annotated[T, ...]` -> `T`
-        if to and is_annotated_type(to):
-            to = extract_type_arg(to, 0)
+        if cast_to and is_annotated_type(cast_to):
+            cast_to = extract_type_arg(cast_to, 0)
 
         if self._is_sse_stream:
             if to:
@@ -164,18 +170,12 @@ def _parse(self, *, to: type[_T] | None = None) -> R | _T:
             return cast(
                 R,
                 stream_cls(
-                    cast_to=self._cast_to,
+                    cast_to=cast_to,
                     response=self.http_response,
                     client=cast(Any, self._client),
                 ),
             )
 
-        cast_to = to if to is not None else self._cast_to
-
-        # unwrap `Annotated[T, ...]` -> `T`
-        if is_annotated_type(cast_to):
-            cast_to = extract_type_arg(cast_to, 0)
-
         if cast_to is NoneType:
             return cast(R, None)
 
@@ -192,6 +192,9 @@ def _parse(self, *, to: type[_T] | None = None) -> R | _T:
         if cast_to == float:
             return cast(R, float(response.text))
 
+        if cast_to == bool:
+            return cast(R, response.text.lower() == "true")
+
         origin = get_origin(cast_to) or cast_to
 
         # handle the legacy binary response case
@@ -315,8 +318,11 @@ class MyModel(BaseModel):
         if is_given(self._options.post_parser):
             parsed = self._options.post_parser(parsed)
 
+        if isinstance(parsed, BaseModel):
+            add_request_id(parsed, self.request_id)
+
         self._parsed_by_type[cache_key] = parsed
-        return parsed
+        return cast(R, parsed)
 
     def read(self) -> bytes:
         """Read and return the binary response content."""
@@ -419,8 +425,11 @@ class MyModel(BaseModel):
         if is_given(self._options.post_parser):
             parsed = self._options.post_parser(parsed)
 
+        if isinstance(parsed, BaseModel):
+            add_request_id(parsed, self.request_id)
+
         self._parsed_by_type[cache_key] = parsed
-        return parsed
+        return cast(R, parsed)
 
     async def read(self) -> bytes:
         """Read and return the binary response content."""
diff --git a/portkey_ai/_vendor/openai/_types.py b/portkey_ai/_vendor/openai/_types.py
index 5611b2d3..a5cf207a 100644
--- a/portkey_ai/_vendor/openai/_types.py
+++ b/portkey_ai/_vendor/openai/_types.py
@@ -16,7 +16,7 @@
     Optional,
     Sequence,
 )
-from typing_extensions import Literal, Protocol, TypeAlias, TypedDict, override, runtime_checkable
+from typing_extensions import Set, Literal, Protocol, TypeAlias, TypedDict, override, runtime_checkable
 
 import httpx
 import pydantic
@@ -194,8 +194,8 @@ def get(self, __key: str) -> str | None: ...
 StrBytesIntFloat = Union[str, bytes, int, float]
 
 # Note: copied from Pydantic
-# https://github.com/pydantic/pydantic/blob/32ea570bf96e84234d2992e1ddf40ab8a565925a/pydantic/main.py#L49
-IncEx: TypeAlias = "set[int] | set[str] | dict[int, Any] | dict[str, Any] | None"
+# https://github.com/pydantic/pydantic/blob/6f31f8f68ef011f84357330186f603ff295312fd/pydantic/main.py#L79
+IncEx: TypeAlias = Union[Set[int], Set[str], Mapping[int, Union["IncEx", bool]], Mapping[str, Union["IncEx", bool]]]
 
 PostParser = Callable[[Any], Any]
 
diff --git a/portkey_ai/_vendor/openai/_utils/__init__.py b/portkey_ai/_vendor/openai/_utils/__init__.py
index 3efe66c8..af2c9bb7 100644
--- a/portkey_ai/_vendor/openai/_utils/__init__.py
+++ b/portkey_ai/_vendor/openai/_utils/__init__.py
@@ -1,3 +1,4 @@
+from ._logs import SensitiveHeadersFilter as SensitiveHeadersFilter
 from ._sync import asyncify as asyncify
 from ._proxy import LazyProxy as LazyProxy
 from ._utils import (
@@ -6,6 +7,7 @@
     is_list as is_list,
     is_given as is_given,
     is_tuple as is_tuple,
+    json_safe as json_safe,
     lru_cache as lru_cache,
     is_mapping as is_mapping,
     is_tuple_t as is_tuple_t,
@@ -38,6 +40,7 @@
     is_iterable_type as is_iterable_type,
     is_required_type as is_required_type,
     is_annotated_type as is_annotated_type,
+    is_type_alias_type as is_type_alias_type,
     strip_annotated_type as strip_annotated_type,
     extract_type_var_from_base as extract_type_var_from_base,
 )
diff --git a/portkey_ai/_vendor/openai/_utils/_logs.py b/portkey_ai/_vendor/openai/_utils/_logs.py
index e5113fd8..37694693 100644
--- a/portkey_ai/_vendor/openai/_utils/_logs.py
+++ b/portkey_ai/_vendor/openai/_utils/_logs.py
@@ -1,10 +1,16 @@
 import os
 import logging
+from typing_extensions import override
+
+from ._utils import is_dict
 
 logger: logging.Logger = logging.getLogger("openai")
 httpx_logger: logging.Logger = logging.getLogger("httpx")
 
 
+SENSITIVE_HEADERS = {"api-key", "authorization"}
+
+
 def _basic_config() -> None:
     # e.g. [2023-10-05 14:12:26 - openai._base_client:818 - DEBUG] HTTP Request: POST http://127.0.0.1:4010/foo/bar "200 OK"
     logging.basicConfig(
@@ -23,3 +29,14 @@ def setup_logging() -> None:
         _basic_config()
         logger.setLevel(logging.INFO)
         httpx_logger.setLevel(logging.INFO)
+
+
+class SensitiveHeadersFilter(logging.Filter):
+    @override
+    def filter(self, record: logging.LogRecord) -> bool:
+        if is_dict(record.args) and "headers" in record.args and is_dict(record.args["headers"]):
+            headers = record.args["headers"] = {**record.args["headers"]}
+            for header in headers:
+                if str(header).lower() in SENSITIVE_HEADERS:
+                    headers[header] = "<redacted>"
+        return True
diff --git a/portkey_ai/_vendor/openai/_utils/_reflection.py b/portkey_ai/_vendor/openai/_utils/_reflection.py
index 89aa712a..bdaca29e 100644
--- a/portkey_ai/_vendor/openai/_utils/_reflection.py
+++ b/portkey_ai/_vendor/openai/_utils/_reflection.py
@@ -15,6 +15,7 @@ def assert_signatures_in_sync(
     check_func: Callable[..., Any],
     *,
     exclude_params: set[str] = set(),
+    description: str = "",
 ) -> None:
     """Ensure that the signature of the second function matches the first."""
 
@@ -39,4 +40,6 @@ def assert_signatures_in_sync(
             continue
 
     if errors:
-        raise AssertionError(f"{len(errors)} errors encountered when comparing signatures:\n\n" + "\n\n".join(errors))
+        raise AssertionError(
+            f"{len(errors)} errors encountered when comparing signatures{description}:\n\n" + "\n\n".join(errors)
+        )
diff --git a/portkey_ai/_vendor/openai/_utils/_sync.py b/portkey_ai/_vendor/openai/_utils/_sync.py
index d0d81033..5d9e2c2a 100644
--- a/portkey_ai/_vendor/openai/_utils/_sync.py
+++ b/portkey_ai/_vendor/openai/_utils/_sync.py
@@ -1,56 +1,61 @@
 from __future__ import annotations
 
+import sys
+import asyncio
 import functools
-from typing import TypeVar, Callable, Awaitable
+import contextvars
+from typing import Any, TypeVar, Callable, Awaitable
 from typing_extensions import ParamSpec
 
-import anyio
-import anyio.to_thread
-
-from ._reflection import function_has_argument
-
 T_Retval = TypeVar("T_Retval")
 T_ParamSpec = ParamSpec("T_ParamSpec")
 
 
-# copied from `asyncer`, https://github.com/tiangolo/asyncer
-def asyncify(
-    function: Callable[T_ParamSpec, T_Retval],
-    *,
-    cancellable: bool = False,
-    limiter: anyio.CapacityLimiter | None = None,
-) -> Callable[T_ParamSpec, Awaitable[T_Retval]]:
-    """
-    Take a blocking function and create an async one that receives the same
-    positional and keyword arguments, and that when called, calls the original function
-    in a worker thread using `anyio.to_thread.run_sync()`. Internally,
-    `asyncer.asyncify()` uses the same `anyio.to_thread.run_sync()`, but it supports
-    keyword arguments additional to positional arguments and it adds better support for
-    autocompletion and inline errors for the arguments of the function called and the
-    return value.
+if sys.version_info >= (3, 9):
+    to_thread = asyncio.to_thread
+else:
+    # backport of https://docs.python.org/3/library/asyncio-task.html#asyncio.to_thread
+    # for Python 3.8 support
+    async def to_thread(
+        func: Callable[T_ParamSpec, T_Retval], /, *args: T_ParamSpec.args, **kwargs: T_ParamSpec.kwargs
+    ) -> Any:
+        """Asynchronously run function *func* in a separate thread.
 
-    If the `cancellable` option is enabled and the task waiting for its completion is
-    cancelled, the thread will still run its course but its return value (or any raised
-    exception) will be ignored.
+        Any *args and **kwargs supplied for this function are directly passed
+        to *func*. Also, the current :class:`contextvars.Context` is propagated,
+        allowing context variables from the main thread to be accessed in the
+        separate thread.
 
-    Use it like this:
+        Returns a coroutine that can be awaited to get the eventual result of *func*.
+        """
+        loop = asyncio.events.get_running_loop()
+        ctx = contextvars.copy_context()
+        func_call = functools.partial(ctx.run, func, *args, **kwargs)
+        return await loop.run_in_executor(None, func_call)
+
+
+# inspired by `asyncer`, https://github.com/tiangolo/asyncer
+def asyncify(function: Callable[T_ParamSpec, T_Retval]) -> Callable[T_ParamSpec, Awaitable[T_Retval]]:
+    """
+    Take a blocking function and create an async one that receives the same
+    positional and keyword arguments. For python version 3.9 and above, it uses
+    asyncio.to_thread to run the function in a separate thread. For python version
+    3.8, it uses locally defined copy of the asyncio.to_thread function which was
+    introduced in python 3.9.
 
-    ```Python
-    def do_work(arg1, arg2, kwarg1="", kwarg2="") -> str:
-        # Do work
-        return "Some result"
+    Usage:
 
+    ```python
+    def blocking_func(arg1, arg2, kwarg1=None):
+        # blocking code
+        return result
 
-    result = await to_thread.asyncify(do_work)("spam", "ham", kwarg1="a", kwarg2="b")
-    print(result)
+    result = asyncify(blocking_function)(arg1, arg2, kwarg1=value1)
     ```
 
     ## Arguments
 
     `function`: a blocking regular callable (e.g. a function)
-    `cancellable`: `True` to allow cancellation of the operation
-    `limiter`: capacity limiter to use to limit the total amount of threads running
-        (if omitted, the default limiter is used)
 
     ## Return
 
@@ -60,22 +65,6 @@ def do_work(arg1, arg2, kwarg1="", kwarg2="") -> str:
     """
 
     async def wrapper(*args: T_ParamSpec.args, **kwargs: T_ParamSpec.kwargs) -> T_Retval:
-        partial_f = functools.partial(function, *args, **kwargs)
-
-        # In `v4.1.0` anyio added the `abandon_on_cancel` argument and deprecated the old
-        # `cancellable` argument, so we need to use the new `abandon_on_cancel` to avoid
-        # surfacing deprecation warnings.
-        if function_has_argument(anyio.to_thread.run_sync, "abandon_on_cancel"):
-            return await anyio.to_thread.run_sync(
-                partial_f,
-                abandon_on_cancel=cancellable,
-                limiter=limiter,
-            )
-
-        return await anyio.to_thread.run_sync(
-            partial_f,
-            cancellable=cancellable,
-            limiter=limiter,
-        )
+        return await to_thread(function, *args, **kwargs)
 
     return wrapper
diff --git a/portkey_ai/_vendor/openai/_utils/_transform.py b/portkey_ai/_vendor/openai/_utils/_transform.py
index 47e262a5..a6b62cad 100644
--- a/portkey_ai/_vendor/openai/_utils/_transform.py
+++ b/portkey_ai/_vendor/openai/_utils/_transform.py
@@ -173,6 +173,11 @@ def _transform_recursive(
         # Iterable[T]
         or (is_iterable_type(stripped_type) and is_iterable(data) and not isinstance(data, str))
     ):
+        # dicts are technically iterable, but it is an iterable on the keys of the dict and is not usually
+        # intended as an iterable, so we don't transform it.
+        if isinstance(data, dict):
+            return cast(object, data)
+
         inner_type = extract_type_arg(stripped_type, 0)
         return [_transform_recursive(d, annotation=annotation, inner_type=inner_type) for d in data]
 
@@ -186,7 +191,7 @@ def _transform_recursive(
         return data
 
     if isinstance(data, pydantic.BaseModel):
-        return model_dump(data, exclude_unset=True)
+        return model_dump(data, exclude_unset=True, mode="json")
 
     annotated_type = _get_annotated_type(annotation)
     if annotated_type is None:
@@ -311,6 +316,11 @@ async def _async_transform_recursive(
         # Iterable[T]
         or (is_iterable_type(stripped_type) and is_iterable(data) and not isinstance(data, str))
     ):
+        # dicts are technically iterable, but it is an iterable on the keys of the dict and is not usually
+        # intended as an iterable, so we don't transform it.
+        if isinstance(data, dict):
+            return cast(object, data)
+
         inner_type = extract_type_arg(stripped_type, 0)
         return [await _async_transform_recursive(d, annotation=annotation, inner_type=inner_type) for d in data]
 
@@ -324,7 +334,7 @@ async def _async_transform_recursive(
         return data
 
     if isinstance(data, pydantic.BaseModel):
-        return model_dump(data, exclude_unset=True)
+        return model_dump(data, exclude_unset=True, mode="json")
 
     annotated_type = _get_annotated_type(annotation)
     if annotated_type is None:
diff --git a/portkey_ai/_vendor/openai/_utils/_typing.py b/portkey_ai/_vendor/openai/_utils/_typing.py
index c036991f..278749b1 100644
--- a/portkey_ai/_vendor/openai/_utils/_typing.py
+++ b/portkey_ai/_vendor/openai/_utils/_typing.py
@@ -1,8 +1,17 @@
 from __future__ import annotations
 
+import sys
+import typing
+import typing_extensions
 from typing import Any, TypeVar, Iterable, cast
 from collections import abc as _c_abc
-from typing_extensions import Required, Annotated, get_args, get_origin
+from typing_extensions import (
+    TypeIs,
+    Required,
+    Annotated,
+    get_args,
+    get_origin,
+)
 
 from .._types import InheritsGeneric
 from .._compat import is_union as _is_union
@@ -36,6 +45,26 @@ def is_typevar(typ: type) -> bool:
     return type(typ) == TypeVar  # type: ignore
 
 
+_TYPE_ALIAS_TYPES: tuple[type[typing_extensions.TypeAliasType], ...] = (typing_extensions.TypeAliasType,)
+if sys.version_info >= (3, 12):
+    _TYPE_ALIAS_TYPES = (*_TYPE_ALIAS_TYPES, typing.TypeAliasType)
+
+
+def is_type_alias_type(tp: Any, /) -> TypeIs[typing_extensions.TypeAliasType]:
+    """Return whether the provided argument is an instance of `TypeAliasType`.
+
+    ```python
+    type Int = int
+    is_type_alias_type(Int)
+    # > True
+    Str = TypeAliasType("Str", str)
+    is_type_alias_type(Str)
+    # > True
+    ```
+    """
+    return isinstance(tp, _TYPE_ALIAS_TYPES)
+
+
 # Extracts T from Annotated[T, ...] or from Required[Annotated[T, ...]]
 def strip_annotated_type(typ: type) -> type:
     if is_required_type(typ) or is_annotated_type(typ):
diff --git a/portkey_ai/_vendor/openai/_utils/_utils.py b/portkey_ai/_vendor/openai/_utils/_utils.py
index 2fc5a1c6..e5811bba 100644
--- a/portkey_ai/_vendor/openai/_utils/_utils.py
+++ b/portkey_ai/_vendor/openai/_utils/_utils.py
@@ -16,6 +16,7 @@
     overload,
 )
 from pathlib import Path
+from datetime import date, datetime
 from typing_extensions import TypeGuard
 
 import sniffio
@@ -363,12 +364,13 @@ def file_from_path(path: str) -> FileTypes:
 
 def get_required_header(headers: HeadersLike, header: str) -> str:
     lower_header = header.lower()
-    if isinstance(headers, Mapping):
-        for k, v in headers.items():
+    if is_mapping_t(headers):
+        # mypy doesn't understand the type narrowing here
+        for k, v in headers.items():  # type: ignore
             if k.lower() == lower_header and isinstance(v, str):
                 return v
 
-    """ to deal with the case where the header looks like Stainless-Event-Id """
+    # to deal with the case where the header looks like Stainless-Event-Id
     intercaps_header = re.sub(r"([^\w])(\w)", lambda pat: pat.group(1) + pat.group(2).upper(), header.capitalize())
 
     for normalized_header in [header, lower_header, header.upper(), intercaps_header]:
@@ -394,3 +396,19 @@ def lru_cache(*, maxsize: int | None = 128) -> Callable[[CallableT], CallableT]:
         maxsize=maxsize,
     )
     return cast(Any, wrapper)  # type: ignore[no-any-return]
+
+
+def json_safe(data: object) -> object:
+    """Translates a mapping / sequence recursively in the same fashion
+    as `pydantic` v2's `model_dump(mode="json")`.
+    """
+    if is_mapping(data):
+        return {json_safe(key): json_safe(value) for key, value in data.items()}
+
+    if is_iterable(data) and not isinstance(data, (str, bytes, bytearray)):
+        return [json_safe(item) for item in data]
+
+    if isinstance(data, (datetime, date)):
+        return data.isoformat()
+
+    return data
diff --git a/portkey_ai/_vendor/openai/_version.py b/portkey_ai/_vendor/openai/_version.py
index f88b8dea..c08e68e1 100644
--- a/portkey_ai/_vendor/openai/_version.py
+++ b/portkey_ai/_vendor/openai/_version.py
@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "openai"
-__version__ = "1.40.1"  # x-release-please-version
+__version__ = "1.58.1"  # x-release-please-version
diff --git a/portkey_ai/_vendor/openai/cli/_cli.py b/portkey_ai/_vendor/openai/cli/_cli.py
index f3177938..aa3cd7d1 100644
--- a/portkey_ai/_vendor/openai/cli/_cli.py
+++ b/portkey_ai/_vendor/openai/cli/_cli.py
@@ -15,7 +15,6 @@
 from .. import _ApiType, __version__
 from ._api import register_commands
 from ._utils import can_use_http2
-from .._types import ProxiesDict
 from ._errors import CLIError, display_error
 from .._compat import PYDANTIC_V2, ConfigDict, model_parse
 from .._models import BaseModel
@@ -167,17 +166,17 @@ def _main() -> None:
     if args.verbosity != 0:
         sys.stderr.write("Warning: --verbosity isn't supported yet\n")
 
-    proxies: ProxiesDict = {}
+    proxies: dict[str, httpx.BaseTransport] = {}
     if args.proxy is not None:
         for proxy in args.proxy:
             key = "https://" if proxy.startswith("https") else "http://"
             if key in proxies:
                 raise CLIError(f"Multiple {key} proxies given - only the last one would be used")
 
-            proxies[key] = proxy
+            proxies[key] = httpx.HTTPTransport(proxy=httpx.Proxy(httpx.URL(proxy)))
 
     http_client = httpx.Client(
-        proxies=proxies or None,
+        mounts=proxies or None,
         http2=can_use_http2(),
     )
     openai.http_client = http_client
diff --git a/portkey_ai/_vendor/openai/cli/_tools/migrate.py b/portkey_ai/_vendor/openai/cli/_tools/migrate.py
index 7c10bb7f..841b7775 100644
--- a/portkey_ai/_vendor/openai/cli/_tools/migrate.py
+++ b/portkey_ai/_vendor/openai/cli/_tools/migrate.py
@@ -2,7 +2,6 @@
 
 import os
 import sys
-import json
 import shutil
 import tarfile
 import platform
@@ -85,14 +84,16 @@ def install() -> Path:
     if sys.platform == "win32":
         raise CLIError("Windows is not supported yet in the migration CLI")
 
-    platform = "macos" if sys.platform == "darwin" else "linux"
+    _debug("Using Grit installer from GitHub")
+
+    platform = "apple-darwin" if sys.platform == "darwin" else "unknown-linux-gnu"
 
     dir_name = _cache_dir() / "openai-python"
     install_dir = dir_name / ".install"
     target_dir = install_dir / "bin"
 
-    target_path = target_dir / "marzano"
-    temp_file = target_dir / "marzano.tmp"
+    target_path = target_dir / "grit"
+    temp_file = target_dir / "grit.tmp"
 
     if target_path.exists():
         _debug(f"{target_path} already exists")
@@ -109,27 +110,14 @@ def install() -> Path:
     arch = _get_arch()
     _debug(f"Using architecture {arch}")
 
-    file_name = f"marzano-{platform}-{arch}"
-    meta_url = f"https://api.keygen.sh/v1/accounts/{KEYGEN_ACCOUNT}/artifacts/{file_name}"
+    file_name = f"grit-{arch}-{platform}"
+    download_url = f"https://github.com/getgrit/gritql/releases/latest/download/{file_name}.tar.gz"
 
-    sys.stdout.write(f"Retrieving Grit CLI metadata from {meta_url}\n")
+    sys.stdout.write(f"Downloading Grit CLI from {download_url}\n")
     with httpx.Client() as client:
-        response = client.get(meta_url)  # pyright: ignore[reportUnknownMemberType]
-
-        data = response.json()
-        errors = data.get("errors")
-        if errors:
-            for error in errors:
-                sys.stdout.write(f"{error}\n")
-
-            raise CLIError("Could not locate Grit CLI binary - see above errors")
-
-        write_manifest(install_dir, data["data"]["relationships"]["release"]["data"]["id"])
-
-        link = data["data"]["links"]["redirect"]
-        _debug(f"Redirect URL {link}")
-
-        download_response = client.get(link)  # pyright: ignore[reportUnknownMemberType]
+        download_response = client.get(download_url, follow_redirects=True)
+        if download_response.status_code != 200:
+            raise CLIError(f"Failed to download Grit CLI from {download_url}")
         with open(temp_file, "wb") as file:
             for chunk in download_response.iter_bytes():
                 file.write(chunk)
@@ -143,8 +131,7 @@ def install() -> Path:
         else:
             archive.extractall(unpacked_dir)
 
-    for item in unpacked_dir.iterdir():
-        item.rename(target_dir / item.name)
+    _move_files_recursively(unpacked_dir, target_dir)
 
     shutil.rmtree(unpacked_dir)
     os.remove(temp_file)
@@ -155,30 +142,23 @@ def install() -> Path:
     return target_path
 
 
+def _move_files_recursively(source_dir: Path, target_dir: Path) -> None:
+    for item in source_dir.iterdir():
+        if item.is_file():
+            item.rename(target_dir / item.name)
+        elif item.is_dir():
+            _move_files_recursively(item, target_dir)
+
+
 def _get_arch() -> str:
     architecture = platform.machine().lower()
 
-    # Map the architecture names to Node.js equivalents
+    # Map the architecture names to Grit equivalents
     arch_map = {
-        "x86_64": "x64",
-        "amd64": "x64",
-        "armv7l": "arm",
-        "aarch64": "arm64",
+        "x86_64": "x86_64",
+        "amd64": "x86_64",
+        "armv7l": "aarch64",
+        "arm64": "aarch64",
     }
 
     return arch_map.get(architecture, architecture)
-
-
-def write_manifest(install_path: Path, release: str) -> None:
-    manifest = {
-        "installPath": str(install_path),
-        "binaries": {
-            "marzano": {
-                "name": "marzano",
-                "release": release,
-            },
-        },
-    }
-    manifest_path = Path(install_path) / "manifests.json"
-    with open(manifest_path, "w") as f:
-        json.dump(manifest, f, indent=2)
diff --git a/portkey_ai/_vendor/openai/lib/_parsing/_completions.py b/portkey_ai/_vendor/openai/lib/_parsing/_completions.py
index f9d1d6b3..f1fa9f2b 100644
--- a/portkey_ai/_vendor/openai/lib/_parsing/_completions.py
+++ b/portkey_ai/_vendor/openai/lib/_parsing/_completions.py
@@ -9,9 +9,9 @@
 from .._tools import PydanticFunctionTool
 from ..._types import NOT_GIVEN, NotGiven
 from ..._utils import is_dict, is_given
-from ..._compat import model_parse_json
+from ..._compat import PYDANTIC_V2, model_parse_json
 from ..._models import construct_type_unchecked
-from .._pydantic import to_strict_json_schema
+from .._pydantic import is_basemodel_type, to_strict_json_schema, is_dataclass_like_type
 from ...types.chat import (
     ParsedChoice,
     ChatCompletion,
@@ -69,7 +69,7 @@ def parse_chat_completion(
     choices: list[ParsedChoice[ResponseFormatT]] = []
     for choice in chat_completion.choices:
         if choice.finish_reason == "length":
-            raise LengthFinishReasonError()
+            raise LengthFinishReasonError(completion=chat_completion)
 
         if choice.finish_reason == "content_filter":
             raise ContentFilterFinishReasonError()
@@ -216,14 +216,16 @@ def is_parseable_tool(input_tool: ChatCompletionToolParam) -> bool:
     return cast(FunctionDefinition, input_fn).get("strict") or False
 
 
-def is_basemodel_type(typ: type) -> TypeGuard[type[pydantic.BaseModel]]:
-    return issubclass(typ, pydantic.BaseModel)
-
-
 def _parse_content(response_format: type[ResponseFormatT], content: str) -> ResponseFormatT:
     if is_basemodel_type(response_format):
         return cast(ResponseFormatT, model_parse_json(response_format, content))
 
+    if is_dataclass_like_type(response_format):
+        if not PYDANTIC_V2:
+            raise TypeError(f"Non BaseModel types are only supported with Pydantic v2 - {response_format}")
+
+        return pydantic.TypeAdapter(response_format).validate_json(content)
+
     raise TypeError(f"Unable to automatically parse response format type {response_format}")
 
 
@@ -241,14 +243,22 @@ def type_to_response_format_param(
     # can only be a `type`
     response_format = cast(type, response_format)
 
-    if not is_basemodel_type(response_format):
+    json_schema_type: type[pydantic.BaseModel] | pydantic.TypeAdapter[Any] | None = None
+
+    if is_basemodel_type(response_format):
+        name = response_format.__name__
+        json_schema_type = response_format
+    elif is_dataclass_like_type(response_format):
+        name = response_format.__name__
+        json_schema_type = pydantic.TypeAdapter(response_format)
+    else:
         raise TypeError(f"Unsupported response_format type - {response_format}")
 
     return {
         "type": "json_schema",
         "json_schema": {
-            "schema": to_strict_json_schema(response_format),
-            "name": response_format.__name__,
+            "schema": to_strict_json_schema(json_schema_type),
+            "name": name,
             "strict": True,
         },
     }
diff --git a/portkey_ai/_vendor/openai/lib/_pydantic.py b/portkey_ai/_vendor/openai/lib/_pydantic.py
index 967ad5de..22c7a1f3 100644
--- a/portkey_ai/_vendor/openai/lib/_pydantic.py
+++ b/portkey_ai/_vendor/openai/lib/_pydantic.py
@@ -1,21 +1,34 @@
 from __future__ import annotations
 
-from typing import Any
+import inspect
+from typing import Any, TypeVar
 from typing_extensions import TypeGuard
 
 import pydantic
 
+from .._types import NOT_GIVEN
 from .._utils import is_dict as _is_dict, is_list
-from .._compat import model_json_schema
+from .._compat import PYDANTIC_V2, model_json_schema
 
+_T = TypeVar("_T")
 
-def to_strict_json_schema(model: type[pydantic.BaseModel]) -> dict[str, Any]:
-    return _ensure_strict_json_schema(model_json_schema(model), path=())
+
+def to_strict_json_schema(model: type[pydantic.BaseModel] | pydantic.TypeAdapter[Any]) -> dict[str, Any]:
+    if inspect.isclass(model) and is_basemodel_type(model):
+        schema = model_json_schema(model)
+    elif PYDANTIC_V2 and isinstance(model, pydantic.TypeAdapter):
+        schema = model.json_schema()
+    else:
+        raise TypeError(f"Non BaseModel types are only supported with Pydantic v2 - {model}")
+
+    return _ensure_strict_json_schema(schema, path=(), root=schema)
 
 
 def _ensure_strict_json_schema(
     json_schema: object,
+    *,
     path: tuple[str, ...],
+    root: dict[str, object],
 ) -> dict[str, Any]:
     """Mutates the given JSON schema to ensure it conforms to the `strict` standard
     that the API expects.
@@ -23,6 +36,16 @@ def _ensure_strict_json_schema(
     if not is_dict(json_schema):
         raise TypeError(f"Expected {json_schema} to be a dictionary; path={path}")
 
+    defs = json_schema.get("$defs")
+    if is_dict(defs):
+        for def_name, def_schema in defs.items():
+            _ensure_strict_json_schema(def_schema, path=(*path, "$defs", def_name), root=root)
+
+    definitions = json_schema.get("definitions")
+    if is_dict(definitions):
+        for definition_name, definition_schema in definitions.items():
+            _ensure_strict_json_schema(definition_schema, path=(*path, "definitions", definition_name), root=root)
+
     typ = json_schema.get("type")
     if typ == "object" and "additionalProperties" not in json_schema:
         json_schema["additionalProperties"] = False
@@ -33,7 +56,7 @@ def _ensure_strict_json_schema(
     if is_dict(properties):
         json_schema["required"] = [prop for prop in properties.keys()]
         json_schema["properties"] = {
-            key: _ensure_strict_json_schema(prop_schema, path=(*path, "properties", key))
+            key: _ensure_strict_json_schema(prop_schema, path=(*path, "properties", key), root=root)
             for key, prop_schema in properties.items()
         }
 
@@ -41,31 +64,87 @@ def _ensure_strict_json_schema(
     # { 'type': 'array', 'items': {...} }
     items = json_schema.get("items")
     if is_dict(items):
-        json_schema["items"] = _ensure_strict_json_schema(items, path=(*path, "items"))
+        json_schema["items"] = _ensure_strict_json_schema(items, path=(*path, "items"), root=root)
 
     # unions
     any_of = json_schema.get("anyOf")
     if is_list(any_of):
         json_schema["anyOf"] = [
-            _ensure_strict_json_schema(variant, path=(*path, "anyOf", str(i))) for i, variant in enumerate(any_of)
+            _ensure_strict_json_schema(variant, path=(*path, "anyOf", str(i)), root=root)
+            for i, variant in enumerate(any_of)
         ]
 
     # intersections
     all_of = json_schema.get("allOf")
     if is_list(all_of):
-        json_schema["allOf"] = [
-            _ensure_strict_json_schema(entry, path=(*path, "anyOf", str(i))) for i, entry in enumerate(all_of)
-        ]
-
-    defs = json_schema.get("$defs")
-    if is_dict(defs):
-        for def_name, def_schema in defs.items():
-            _ensure_strict_json_schema(def_schema, path=(*path, "$defs", def_name))
+        if len(all_of) == 1:
+            json_schema.update(_ensure_strict_json_schema(all_of[0], path=(*path, "allOf", "0"), root=root))
+            json_schema.pop("allOf")
+        else:
+            json_schema["allOf"] = [
+                _ensure_strict_json_schema(entry, path=(*path, "allOf", str(i)), root=root)
+                for i, entry in enumerate(all_of)
+            ]
+
+    # strip `None` defaults as there's no meaningful distinction here
+    # the schema will still be `nullable` and the model will default
+    # to using `None` anyway
+    if json_schema.get("default", NOT_GIVEN) is None:
+        json_schema.pop("default")
+
+    # we can't use `$ref`s if there are also other properties defined, e.g.
+    # `{"$ref": "...", "description": "my description"}`
+    #
+    # so we unravel the ref
+    # `{"type": "string", "description": "my description"}`
+    ref = json_schema.get("$ref")
+    if ref and has_more_than_n_keys(json_schema, 1):
+        assert isinstance(ref, str), f"Received non-string $ref - {ref}"
+
+        resolved = resolve_ref(root=root, ref=ref)
+        if not is_dict(resolved):
+            raise ValueError(f"Expected `$ref: {ref}` to resolved to a dictionary but got {resolved}")
+
+        # properties from the json schema take priority over the ones on the `$ref`
+        json_schema.update({**resolved, **json_schema})
+        json_schema.pop("$ref")
 
     return json_schema
 
 
+def resolve_ref(*, root: dict[str, object], ref: str) -> object:
+    if not ref.startswith("#/"):
+        raise ValueError(f"Unexpected $ref format {ref!r}; Does not start with #/")
+
+    path = ref[2:].split("/")
+    resolved = root
+    for key in path:
+        value = resolved[key]
+        assert is_dict(value), f"encountered non-dictionary entry while resolving {ref} - {resolved}"
+        resolved = value
+
+    return resolved
+
+
+def is_basemodel_type(typ: type) -> TypeGuard[type[pydantic.BaseModel]]:
+    return issubclass(typ, pydantic.BaseModel)
+
+
+def is_dataclass_like_type(typ: type) -> bool:
+    """Returns True if the given type likely used `@pydantic.dataclass`"""
+    return hasattr(typ, "__pydantic_config__")
+
+
 def is_dict(obj: object) -> TypeGuard[dict[str, object]]:
     # just pretend that we know there are only `str` keys
     # as that check is not worth the performance cost
     return _is_dict(obj)
+
+
+def has_more_than_n_keys(obj: dict[str, object], n: int) -> bool:
+    i = 0
+    for _ in obj.keys():
+        i += 1
+        if i > n:
+            return True
+    return False
diff --git a/portkey_ai/_vendor/openai/lib/azure.py b/portkey_ai/_vendor/openai/lib/azure.py
index ef64137d..13d9f318 100644
--- a/portkey_ai/_vendor/openai/lib/azure.py
+++ b/portkey_ai/_vendor/openai/lib/azure.py
@@ -53,13 +53,15 @@ class BaseAzureClient(BaseClient[_HttpxClientT, _DefaultStreamT]):
     def _build_request(
         self,
         options: FinalRequestOptions,
+        *,
+        retries_taken: int = 0,
     ) -> httpx.Request:
         if options.url in _deployments_endpoints and is_mapping(options.json_data):
             model = options.json_data.get("model")
             if model is not None and not "/deployments" in str(self.base_url):
                 options.url = f"/deployments/{model}{options.url}"
 
-        return super()._build_request(options)
+        return super()._build_request(options, retries_taken=retries_taken)
 
 
 class AzureOpenAI(BaseAzureClient[httpx.Client, Stream[Any]], OpenAI):
@@ -74,6 +76,7 @@ def __init__(
         azure_ad_token: str | None = None,
         azure_ad_token_provider: AzureADTokenProvider | None = None,
         organization: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
         default_headers: Mapping[str, str] | None = None,
@@ -92,6 +95,7 @@ def __init__(
         azure_ad_token: str | None = None,
         azure_ad_token_provider: AzureADTokenProvider | None = None,
         organization: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
         default_headers: Mapping[str, str] | None = None,
@@ -110,6 +114,7 @@ def __init__(
         azure_ad_token: str | None = None,
         azure_ad_token_provider: AzureADTokenProvider | None = None,
         organization: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
         default_headers: Mapping[str, str] | None = None,
@@ -129,6 +134,7 @@ def __init__(
         azure_ad_token_provider: AzureADTokenProvider | None = None,
         organization: str | None = None,
         project: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         base_url: str | None = None,
         timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
@@ -191,9 +197,9 @@ def __init__(
                 )
 
             if azure_deployment is not None:
-                base_url = f"{azure_endpoint}/openai/deployments/{azure_deployment}"
+                base_url = f"{azure_endpoint.rstrip('/')}/openai/deployments/{azure_deployment}"
             else:
-                base_url = f"{azure_endpoint}/openai"
+                base_url = f"{azure_endpoint.rstrip('/')}/openai"
         else:
             if azure_endpoint is not None:
                 raise ValueError("base_url and azure_endpoint are mutually exclusive")
@@ -212,6 +218,7 @@ def __init__(
             default_headers=default_headers,
             default_query=default_query,
             http_client=http_client,
+            websocket_base_url=websocket_base_url,
             _strict_response_validation=_strict_response_validation,
         )
         self._api_version = api_version
@@ -225,6 +232,7 @@ def copy(
         api_key: str | None = None,
         organization: str | None = None,
         project: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         api_version: str | None = None,
         azure_ad_token: str | None = None,
         azure_ad_token_provider: AzureADTokenProvider | None = None,
@@ -245,6 +253,7 @@ def copy(
             api_key=api_key,
             organization=organization,
             project=project,
+            websocket_base_url=websocket_base_url,
             base_url=base_url,
             timeout=timeout,
             http_client=http_client,
@@ -312,6 +321,7 @@ def __init__(
         azure_ad_token_provider: AsyncAzureADTokenProvider | None = None,
         organization: str | None = None,
         project: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
         default_headers: Mapping[str, str] | None = None,
@@ -331,6 +341,7 @@ def __init__(
         azure_ad_token_provider: AsyncAzureADTokenProvider | None = None,
         organization: str | None = None,
         project: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
         default_headers: Mapping[str, str] | None = None,
@@ -350,6 +361,7 @@ def __init__(
         azure_ad_token_provider: AsyncAzureADTokenProvider | None = None,
         organization: str | None = None,
         project: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
         default_headers: Mapping[str, str] | None = None,
@@ -370,6 +382,7 @@ def __init__(
         organization: str | None = None,
         project: str | None = None,
         base_url: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         timeout: float | Timeout | None | NotGiven = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
         default_headers: Mapping[str, str] | None = None,
@@ -431,9 +444,9 @@ def __init__(
                 )
 
             if azure_deployment is not None:
-                base_url = f"{azure_endpoint}/openai/deployments/{azure_deployment}"
+                base_url = f"{azure_endpoint.rstrip('/')}/openai/deployments/{azure_deployment}"
             else:
-                base_url = f"{azure_endpoint}/openai"
+                base_url = f"{azure_endpoint.rstrip('/')}/openai"
         else:
             if azure_endpoint is not None:
                 raise ValueError("base_url and azure_endpoint are mutually exclusive")
@@ -452,6 +465,7 @@ def __init__(
             default_headers=default_headers,
             default_query=default_query,
             http_client=http_client,
+            websocket_base_url=websocket_base_url,
             _strict_response_validation=_strict_response_validation,
         )
         self._api_version = api_version
@@ -465,6 +479,7 @@ def copy(
         api_key: str | None = None,
         organization: str | None = None,
         project: str | None = None,
+        websocket_base_url: str | httpx.URL | None = None,
         api_version: str | None = None,
         azure_ad_token: str | None = None,
         azure_ad_token_provider: AsyncAzureADTokenProvider | None = None,
@@ -485,6 +500,7 @@ def copy(
             api_key=api_key,
             organization=organization,
             project=project,
+            websocket_base_url=websocket_base_url,
             base_url=base_url,
             timeout=timeout,
             http_client=http_client,
diff --git a/portkey_ai/_vendor/openai/lib/streaming/_assistants.py b/portkey_ai/_vendor/openai/lib/streaming/_assistants.py
index 7445f9a9..6efb3ca3 100644
--- a/portkey_ai/_vendor/openai/lib/streaming/_assistants.py
+++ b/portkey_ai/_vendor/openai/lib/streaming/_assistants.py
@@ -8,6 +8,7 @@
 import httpx
 
 from ..._utils import is_dict, is_list, consume_sync_iterator, consume_async_iterator
+from ..._compat import model_dump
 from ..._models import construct_type
 from ..._streaming import Stream, AsyncStream
 from ...types.beta import AssistantStreamEvent
@@ -242,7 +243,7 @@ def on_text_delta(self, delta: TextDelta, snapshot: Text) -> None:
         on_text_delta(TextDelta(value=" solution"), Text(value="The solution")),
         on_text_delta(TextDelta(value=" to"), Text(value="The solution to")),
         on_text_delta(TextDelta(value=" the"), Text(value="The solution to the")),
-        on_text_delta(TextDelta(value=" equation"), Text(value="The solution to the equivalent")),
+        on_text_delta(TextDelta(value=" equation"), Text(value="The solution to the equation")),
         """
 
     def on_text_done(self, text: Text) -> None:
@@ -906,11 +907,11 @@ def accumulate_run_step(
             merged = accumulate_delta(
                 cast(
                     "dict[object, object]",
-                    snapshot.model_dump(exclude_unset=True),
+                    model_dump(snapshot, exclude_unset=True, warnings=False),
                 ),
                 cast(
                     "dict[object, object]",
-                    data.delta.model_dump(exclude_unset=True),
+                    model_dump(data.delta, exclude_unset=True, warnings=False),
                 ),
             )
             run_step_snapshots[snapshot.id] = cast(RunStep, construct_type(type_=RunStep, value=merged))
@@ -948,7 +949,7 @@ def accumulate_event(
                         construct_type(
                             # mypy doesn't allow Content for some reason
                             type_=cast(Any, MessageContent),
-                            value=content_delta.model_dump(exclude_unset=True),
+                            value=model_dump(content_delta, exclude_unset=True, warnings=False),
                         ),
                     ),
                 )
@@ -957,11 +958,11 @@ def accumulate_event(
                 merged = accumulate_delta(
                     cast(
                         "dict[object, object]",
-                        block.model_dump(exclude_unset=True),
+                        model_dump(block, exclude_unset=True, warnings=False),
                     ),
                     cast(
                         "dict[object, object]",
-                        content_delta.model_dump(exclude_unset=True),
+                        model_dump(content_delta, exclude_unset=True, warnings=False),
                     ),
                 )
                 current_message_snapshot.content[content_delta.index] = cast(
diff --git a/portkey_ai/_vendor/openai/lib/streaming/chat/__init__.py b/portkey_ai/_vendor/openai/lib/streaming/chat/__init__.py
index 5881c39b..dfa3f3f2 100644
--- a/portkey_ai/_vendor/openai/lib/streaming/chat/__init__.py
+++ b/portkey_ai/_vendor/openai/lib/streaming/chat/__init__.py
@@ -21,6 +21,7 @@
 from ._completions import (
     ChatCompletionStream as ChatCompletionStream,
     AsyncChatCompletionStream as AsyncChatCompletionStream,
+    ChatCompletionStreamState as ChatCompletionStreamState,
     ChatCompletionStreamManager as ChatCompletionStreamManager,
     AsyncChatCompletionStreamManager as AsyncChatCompletionStreamManager,
 )
diff --git a/portkey_ai/_vendor/openai/lib/streaming/chat/_completions.py b/portkey_ai/_vendor/openai/lib/streaming/chat/_completions.py
index 342a5e2b..21460913 100644
--- a/portkey_ai/_vendor/openai/lib/streaming/chat/_completions.py
+++ b/portkey_ai/_vendor/openai/lib/streaming/chat/_completions.py
@@ -23,7 +23,7 @@
     FunctionToolCallArgumentsDeltaEvent,
 )
 from .._deltas import accumulate_delta
-from ...._types import NOT_GIVEN, NotGiven
+from ...._types import NOT_GIVEN, IncEx, NotGiven
 from ...._utils import is_given, consume_sync_iterator, consume_async_iterator
 from ...._compat import model_dump
 from ...._models import build, construct_type
@@ -287,11 +287,31 @@ async def __aexit__(
 
 
 class ChatCompletionStreamState(Generic[ResponseFormatT]):
+    """Helper class for manually accumulating `ChatCompletionChunk`s into a final `ChatCompletion` object.
+
+    This is useful in cases where you can't always use the `.stream()` method, e.g.
+
+    ```py
+    from openai.lib.streaming.chat import ChatCompletionStreamState
+
+    state = ChatCompletionStreamState()
+
+    stream = client.chat.completions.create(..., stream=True)
+    for chunk in response:
+        state.handle_chunk(chunk)
+
+        # can also access the accumulated `ChatCompletion` mid-stream
+        state.current_completion_snapshot
+
+    print(state.get_final_completion())
+    ```
+    """
+
     def __init__(
         self,
         *,
-        input_tools: Iterable[ChatCompletionToolParam] | NotGiven,
-        response_format: type[ResponseFormatT] | ResponseFormatParam | NotGiven,
+        input_tools: Iterable[ChatCompletionToolParam] | NotGiven = NOT_GIVEN,
+        response_format: type[ResponseFormatT] | ResponseFormatParam | NotGiven = NOT_GIVEN,
     ) -> None:
         self.__current_completion_snapshot: ParsedChatCompletionSnapshot | None = None
         self.__choice_event_states: list[ChoiceEventState] = []
@@ -301,6 +321,11 @@ def __init__(
         self._rich_response_format: type | NotGiven = response_format if inspect.isclass(response_format) else NOT_GIVEN
 
     def get_final_completion(self) -> ParsedChatCompletion[ResponseFormatT]:
+        """Parse the final completion object.
+
+        Note this does not provide any guarantees that the stream has actually finished, you must
+        only call this method when the stream is finished.
+        """
         return parse_chat_completion(
             chat_completion=self.current_completion_snapshot,
             response_format=self._rich_response_format,
@@ -312,8 +337,8 @@ def current_completion_snapshot(self) -> ParsedChatCompletionSnapshot:
         assert self.__current_completion_snapshot is not None
         return self.__current_completion_snapshot
 
-    def handle_chunk(self, chunk: ChatCompletionChunk) -> list[ChatCompletionStreamEvent[ResponseFormatT]]:
-        """Accumulate a new chunk into the snapshot and returns a list of events to yield."""
+    def handle_chunk(self, chunk: ChatCompletionChunk) -> Iterable[ChatCompletionStreamEvent[ResponseFormatT]]:
+        """Accumulate a new chunk into the snapshot and returns an iterable of events to yield."""
         self.__current_completion_snapshot = self._accumulate_chunk(chunk)
 
         return self._build_events(
@@ -352,13 +377,17 @@ def _accumulate_chunk(self, chunk: ChatCompletionChunk) -> ParsedChatCompletionS
                                     # we don't want to serialise / deserialise our custom properties
                                     # as they won't appear in the delta and we don't want to have to
                                     # continuosly reparse the content
-                                    exclude={
-                                        "parsed": True,
-                                        "tool_calls": {
-                                            idx: {"function": {"parsed_arguments": True}}
-                                            for idx, _ in enumerate(choice_snapshot.message.tool_calls or [])
+                                    exclude=cast(
+                                        # cast required as mypy isn't smart enough to infer `True` here to `Literal[True]`
+                                        IncEx,
+                                        {
+                                            "parsed": True,
+                                            "tool_calls": {
+                                                idx: {"function": {"parsed_arguments": True}}
+                                                for idx, _ in enumerate(choice_snapshot.message.tool_calls or [])
+                                            },
                                         },
-                                    },
+                                    ),
                                 ),
                             ),
                             cast("dict[object, object]", choice.delta.to_dict()),
@@ -394,7 +423,9 @@ def _accumulate_chunk(self, chunk: ChatCompletionChunk) -> ParsedChatCompletionS
 
                 if has_parseable_input(response_format=self._response_format, input_tools=self._input_tools):
                     if choice.finish_reason == "length":
-                        raise LengthFinishReasonError()
+                        # at the time of writing, `.usage` will always be `None` but
+                        # we include it here in case that is changed in the future
+                        raise LengthFinishReasonError(completion=completion_snapshot)
 
                     if choice.finish_reason == "content_filter":
                         raise ContentFilterFinishReasonError()
diff --git a/portkey_ai/_vendor/openai/resources/audio/audio.py b/portkey_ai/_vendor/openai/resources/audio/audio.py
index 537ad573..18bd7b81 100644
--- a/portkey_ai/_vendor/openai/resources/audio/audio.py
+++ b/portkey_ai/_vendor/openai/resources/audio/audio.py
@@ -47,10 +47,21 @@ def speech(self) -> Speech:
 
     @cached_property
     def with_raw_response(self) -> AudioWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AudioWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AudioWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AudioWithStreamingResponse(self)
 
 
@@ -69,10 +80,21 @@ def speech(self) -> AsyncSpeech:
 
     @cached_property
     def with_raw_response(self) -> AsyncAudioWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncAudioWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncAudioWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncAudioWithStreamingResponse(self)
 
 
diff --git a/portkey_ai/_vendor/openai/resources/audio/speech.py b/portkey_ai/_vendor/openai/resources/audio/speech.py
index a0df9ec4..09faaddd 100644
--- a/portkey_ai/_vendor/openai/resources/audio/speech.py
+++ b/portkey_ai/_vendor/openai/resources/audio/speech.py
@@ -31,10 +31,21 @@
 class Speech(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> SpeechWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return SpeechWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> SpeechWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return SpeechWithStreamingResponse(self)
 
     def create(
@@ -59,13 +70,13 @@ def create(
           input: The text to generate audio for. The maximum length is 4096 characters.
 
           model:
-              One of the available [TTS models](https://platform.openai.com/docs/models/tts):
+              One of the available [TTS models](https://platform.openai.com/docs/models#tts):
               `tts-1` or `tts-1-hd`
 
           voice: The voice to use when generating the audio. Supported voices are `alloy`,
               `echo`, `fable`, `onyx`, `nova`, and `shimmer`. Previews of the voices are
               available in the
-              [Text to speech guide](https://platform.openai.com/docs/guides/text-to-speech/voice-options).
+              [Text to speech guide](https://platform.openai.com/docs/guides/text-to-speech#voice-options).
 
           response_format: The format to audio in. Supported formats are `mp3`, `opus`, `aac`, `flac`,
               `wav`, and `pcm`.
@@ -104,10 +115,21 @@ def create(
 class AsyncSpeech(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncSpeechWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncSpeechWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncSpeechWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncSpeechWithStreamingResponse(self)
 
     async def create(
@@ -132,13 +154,13 @@ async def create(
           input: The text to generate audio for. The maximum length is 4096 characters.
 
           model:
-              One of the available [TTS models](https://platform.openai.com/docs/models/tts):
+              One of the available [TTS models](https://platform.openai.com/docs/models#tts):
               `tts-1` or `tts-1-hd`
 
           voice: The voice to use when generating the audio. Supported voices are `alloy`,
               `echo`, `fable`, `onyx`, `nova`, and `shimmer`. Previews of the voices are
               available in the
-              [Text to speech guide](https://platform.openai.com/docs/guides/text-to-speech/voice-options).
+              [Text to speech guide](https://platform.openai.com/docs/guides/text-to-speech#voice-options).
 
           response_format: The format to audio in. Supported formats are `mp3`, `opus`, `aac`, `flac`,
               `wav`, and `pcm`.
diff --git a/portkey_ai/_vendor/openai/resources/audio/transcriptions.py b/portkey_ai/_vendor/openai/resources/audio/transcriptions.py
index 1ee96241..8b5f4404 100644
--- a/portkey_ai/_vendor/openai/resources/audio/transcriptions.py
+++ b/portkey_ai/_vendor/openai/resources/audio/transcriptions.py
@@ -2,12 +2,14 @@
 
 from __future__ import annotations
 
-from typing import List, Union, Mapping, cast
-from typing_extensions import Literal
+import logging
+from typing import TYPE_CHECKING, List, Union, Mapping, cast
+from typing_extensions import Literal, overload, assert_never
 
 import httpx
 
 from ... import _legacy_response
+from ...types import AudioResponseFormat
 from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
 from ..._utils import (
     extract_files,
@@ -22,19 +24,91 @@
 from ..._base_client import make_request_options
 from ...types.audio_model import AudioModel
 from ...types.audio.transcription import Transcription
+from ...types.audio_response_format import AudioResponseFormat
+from ...types.audio.transcription_verbose import TranscriptionVerbose
 
 __all__ = ["Transcriptions", "AsyncTranscriptions"]
 
+log: logging.Logger = logging.getLogger("openai.audio.transcriptions")
+
 
 class Transcriptions(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> TranscriptionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return TranscriptionsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> TranscriptionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return TranscriptionsWithStreamingResponse(self)
 
+    @overload
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Union[Literal["json"], NotGiven] = NOT_GIVEN,
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Transcription: ...
+
+    @overload
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["verbose_json"],
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> TranscriptionVerbose: ...
+
+    @overload
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["text", "srt", "vtt"],
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> str: ...
+
     def create(
         self,
         *,
@@ -42,7 +116,7 @@ def create(
         model: Union[str, AudioModel],
         language: str | NotGiven = NOT_GIVEN,
         prompt: str | NotGiven = NOT_GIVEN,
-        response_format: Literal["json", "text", "srt", "verbose_json", "vtt"] | NotGiven = NOT_GIVEN,
+        response_format: Union[AudioResponseFormat, NotGiven] = NOT_GIVEN,
         temperature: float | NotGiven = NOT_GIVEN,
         timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -51,7 +125,7 @@ def create(
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> Transcription:
+    ) -> Transcription | TranscriptionVerbose | str:
         """
         Transcribes audio into the input language.
 
@@ -69,11 +143,11 @@ def create(
 
           prompt: An optional text to guide the model's style or continue a previous audio
               segment. The
-              [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
+              [prompt](https://platform.openai.com/docs/guides/speech-to-text#prompting)
               should match the audio language.
 
-          response_format: The format of the transcript output, in one of these options: `json`, `text`,
-              `srt`, `verbose_json`, or `vtt`.
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
 
           temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
               output more random, while lower values like 0.2 will make it more focused and
@@ -111,34 +185,65 @@ def create(
         # sent to the server will contain a `boundary` parameter, e.g.
         # multipart/form-data; boundary=---abc--
         extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
-        return self._post(
+        return self._post(  # type: ignore[return-value]
             "/audio/transcriptions",
             body=maybe_transform(body, transcription_create_params.TranscriptionCreateParams),
             files=files,
             options=make_request_options(
                 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
             ),
-            cast_to=Transcription,
+            cast_to=_get_response_format_type(response_format),
         )
 
 
 class AsyncTranscriptions(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncTranscriptionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncTranscriptionsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncTranscriptionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncTranscriptionsWithStreamingResponse(self)
 
+    @overload
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Union[Literal["json"], NotGiven] = NOT_GIVEN,
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Transcription: ...
+
+    @overload
     async def create(
         self,
         *,
         file: FileTypes,
         model: Union[str, AudioModel],
+        response_format: Literal["verbose_json"],
         language: str | NotGiven = NOT_GIVEN,
         prompt: str | NotGiven = NOT_GIVEN,
-        response_format: Literal["json", "text", "srt", "verbose_json", "vtt"] | NotGiven = NOT_GIVEN,
         temperature: float | NotGiven = NOT_GIVEN,
         timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -147,7 +252,44 @@ async def create(
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> Transcription:
+    ) -> TranscriptionVerbose: ...
+
+    @overload
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["text", "srt", "vtt"],
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> str: ...
+
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        response_format: Union[AudioResponseFormat, NotGiven] = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Transcription | TranscriptionVerbose | str:
         """
         Transcribes audio into the input language.
 
@@ -165,11 +307,11 @@ async def create(
 
           prompt: An optional text to guide the model's style or continue a previous audio
               segment. The
-              [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
+              [prompt](https://platform.openai.com/docs/guides/speech-to-text#prompting)
               should match the audio language.
 
-          response_format: The format of the transcript output, in one of these options: `json`, `text`,
-              `srt`, `verbose_json`, or `vtt`.
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
 
           temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
               output more random, while lower values like 0.2 will make it more focused and
@@ -214,7 +356,7 @@ async def create(
             options=make_request_options(
                 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
             ),
-            cast_to=Transcription,
+            cast_to=_get_response_format_type(response_format),
         )
 
 
@@ -252,3 +394,22 @@ def __init__(self, transcriptions: AsyncTranscriptions) -> None:
         self.create = async_to_streamed_response_wrapper(
             transcriptions.create,
         )
+
+
+def _get_response_format_type(
+    response_format: Literal["json", "text", "srt", "verbose_json", "vtt"] | NotGiven,
+) -> type[Transcription | TranscriptionVerbose | str]:
+    if isinstance(response_format, NotGiven) or response_format is None:  # pyright: ignore[reportUnnecessaryComparison]
+        return Transcription
+
+    if response_format == "json":
+        return Transcription
+    elif response_format == "verbose_json":
+        return TranscriptionVerbose
+    elif response_format == "srt" or response_format == "text" or response_format == "vtt":
+        return str
+    elif TYPE_CHECKING:  # type: ignore[unreachable]
+        assert_never(response_format)
+    else:
+        log.warn("Unexpected audio response format: %s", response_format)
+        return Transcription
diff --git a/portkey_ai/_vendor/openai/resources/audio/translations.py b/portkey_ai/_vendor/openai/resources/audio/translations.py
index ed97ccf8..a2d28afa 100644
--- a/portkey_ai/_vendor/openai/resources/audio/translations.py
+++ b/portkey_ai/_vendor/openai/resources/audio/translations.py
@@ -2,11 +2,14 @@
 
 from __future__ import annotations
 
-from typing import Union, Mapping, cast
+import logging
+from typing import TYPE_CHECKING, Union, Mapping, cast
+from typing_extensions import Literal, overload, assert_never
 
 import httpx
 
 from ... import _legacy_response
+from ...types import AudioResponseFormat
 from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
 from ..._utils import (
     extract_files,
@@ -21,26 +24,76 @@
 from ..._base_client import make_request_options
 from ...types.audio_model import AudioModel
 from ...types.audio.translation import Translation
+from ...types.audio_response_format import AudioResponseFormat
+from ...types.audio.translation_verbose import TranslationVerbose
 
 __all__ = ["Translations", "AsyncTranslations"]
 
+log: logging.Logger = logging.getLogger("openai.audio.transcriptions")
+
 
 class Translations(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> TranslationsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return TranslationsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> TranslationsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return TranslationsWithStreamingResponse(self)
 
+    @overload
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Union[Literal["json"], NotGiven] = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Translation: ...
+
+    @overload
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["verbose_json"],
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> TranslationVerbose: ...
+
+    @overload
     def create(
         self,
         *,
         file: FileTypes,
         model: Union[str, AudioModel],
+        response_format: Literal["text", "srt", "vtt"],
         prompt: str | NotGiven = NOT_GIVEN,
-        response_format: str | NotGiven = NOT_GIVEN,
         temperature: float | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -48,7 +101,23 @@ def create(
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> Translation:
+    ) -> str: ...
+
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        prompt: str | NotGiven = NOT_GIVEN,
+        response_format: Union[AudioResponseFormat, NotGiven] = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Translation | TranslationVerbose | str:
         """
         Translates audio into English.
 
@@ -61,11 +130,11 @@ def create(
 
           prompt: An optional text to guide the model's style or continue a previous audio
               segment. The
-              [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
+              [prompt](https://platform.openai.com/docs/guides/speech-to-text#prompting)
               should be in English.
 
-          response_format: The format of the transcript output, in one of these options: `json`, `text`,
-              `srt`, `verbose_json`, or `vtt`.
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
 
           temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
               output more random, while lower values like 0.2 will make it more focused and
@@ -95,33 +164,45 @@ def create(
         # sent to the server will contain a `boundary` parameter, e.g.
         # multipart/form-data; boundary=---abc--
         extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
-        return self._post(
+        return self._post(  # type: ignore[return-value]
             "/audio/translations",
             body=maybe_transform(body, translation_create_params.TranslationCreateParams),
             files=files,
             options=make_request_options(
                 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
             ),
-            cast_to=Translation,
+            cast_to=_get_response_format_type(response_format),
         )
 
 
 class AsyncTranslations(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncTranslationsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncTranslationsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncTranslationsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncTranslationsWithStreamingResponse(self)
 
+    @overload
     async def create(
         self,
         *,
         file: FileTypes,
         model: Union[str, AudioModel],
+        response_format: Union[Literal["json"], NotGiven] = NOT_GIVEN,
         prompt: str | NotGiven = NOT_GIVEN,
-        response_format: str | NotGiven = NOT_GIVEN,
         temperature: float | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -129,7 +210,57 @@ async def create(
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> Translation:
+    ) -> Translation: ...
+
+    @overload
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["verbose_json"],
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> TranslationVerbose: ...
+
+    @overload
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["text", "srt", "vtt"],
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> str: ...
+
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        prompt: str | NotGiven = NOT_GIVEN,
+        response_format: Union[AudioResponseFormat, NotGiven] = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Translation | TranslationVerbose | str:
         """
         Translates audio into English.
 
@@ -142,11 +273,11 @@ async def create(
 
           prompt: An optional text to guide the model's style or continue a previous audio
               segment. The
-              [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
+              [prompt](https://platform.openai.com/docs/guides/speech-to-text#prompting)
               should be in English.
 
-          response_format: The format of the transcript output, in one of these options: `json`, `text`,
-              `srt`, `verbose_json`, or `vtt`.
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
 
           temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
               output more random, while lower values like 0.2 will make it more focused and
@@ -183,7 +314,7 @@ async def create(
             options=make_request_options(
                 extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
             ),
-            cast_to=Translation,
+            cast_to=_get_response_format_type(response_format),
         )
 
 
@@ -221,3 +352,22 @@ def __init__(self, translations: AsyncTranslations) -> None:
         self.create = async_to_streamed_response_wrapper(
             translations.create,
         )
+
+
+def _get_response_format_type(
+    response_format: Literal["json", "text", "srt", "verbose_json", "vtt"] | NotGiven,
+) -> type[Translation | TranslationVerbose | str]:
+    if isinstance(response_format, NotGiven) or response_format is None:  # pyright: ignore[reportUnnecessaryComparison]
+        return Translation
+
+    if response_format == "json":
+        return Translation
+    elif response_format == "verbose_json":
+        return TranslationVerbose
+    elif response_format == "srt" or response_format == "text" or response_format == "vtt":
+        return str
+    elif TYPE_CHECKING:  # type: ignore[unreachable]
+        assert_never(response_format)
+    else:
+        log.warn("Unexpected audio response format: %s", response_format)
+        return Transcription
diff --git a/portkey_ai/_vendor/openai/resources/batches.py b/portkey_ai/_vendor/openai/resources/batches.py
index 7152fac6..7cab7578 100644
--- a/portkey_ai/_vendor/openai/resources/batches.py
+++ b/portkey_ai/_vendor/openai/resources/batches.py
@@ -30,10 +30,21 @@
 class Batches(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> BatchesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return BatchesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> BatchesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return BatchesWithStreamingResponse(self)
 
     def create(
@@ -70,7 +81,7 @@ def create(
               Your input file must be formatted as a
               [JSONL file](https://platform.openai.com/docs/api-reference/batch/request-input),
               and must be uploaded with the purpose `batch`. The file can contain up to 50,000
-              requests, and can be up to 100 MB in size.
+              requests, and can be up to 200 MB in size.
 
           metadata: Optional custom metadata for the batch.
 
@@ -224,10 +235,21 @@ def cancel(
 class AsyncBatches(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncBatchesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncBatchesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncBatchesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncBatchesWithStreamingResponse(self)
 
     async def create(
@@ -264,7 +286,7 @@ async def create(
               Your input file must be formatted as a
               [JSONL file](https://platform.openai.com/docs/api-reference/batch/request-input),
               and must be uploaded with the purpose `batch`. The file can contain up to 50,000
-              requests, and can be up to 100 MB in size.
+              requests, and can be up to 200 MB in size.
 
           metadata: Optional custom metadata for the batch.
 
diff --git a/portkey_ai/_vendor/openai/resources/beta/assistants.py b/portkey_ai/_vendor/openai/resources/beta/assistants.py
index 441390d2..7df212f1 100644
--- a/portkey_ai/_vendor/openai/resources/beta/assistants.py
+++ b/portkey_ai/_vendor/openai/resources/beta/assistants.py
@@ -35,10 +35,21 @@
 class Assistants(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AssistantsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AssistantsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AssistantsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AssistantsWithStreamingResponse(self)
 
     def create(
@@ -68,8 +79,8 @@ def create(
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           description: The description of the assistant. The maximum length is 512 characters.
 
@@ -78,22 +89,22 @@ def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           name: The name of the assistant. The maximum length is 256 characters.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -222,28 +233,28 @@ def update(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           name: The name of the assistant. The maximum length is 256 characters.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -333,8 +344,8 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           limit: A limit on the number of objects to be returned. Limit can range between 1 and
               100, and the default is 20.
@@ -410,10 +421,21 @@ def delete(
 class AsyncAssistants(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncAssistantsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncAssistantsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncAssistantsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncAssistantsWithStreamingResponse(self)
 
     async def create(
@@ -443,8 +465,8 @@ async def create(
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           description: The description of the assistant. The maximum length is 512 characters.
 
@@ -453,22 +475,22 @@ async def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           name: The name of the assistant. The maximum length is 256 characters.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -597,28 +619,28 @@ async def update(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           name: The name of the assistant. The maximum length is 256 characters.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -708,8 +730,8 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           limit: A limit on the number of objects to be returned. Limit can range between 1 and
               100, and the default is 20.
diff --git a/portkey_ai/_vendor/openai/resources/beta/beta.py b/portkey_ai/_vendor/openai/resources/beta/beta.py
index 479c97c4..1ffa6c8e 100644
--- a/portkey_ai/_vendor/openai/resources/beta/beta.py
+++ b/portkey_ai/_vendor/openai/resources/beta/beta.py
@@ -2,14 +2,6 @@
 
 from __future__ import annotations
 
-from .threads import (
-    Threads,
-    AsyncThreads,
-    ThreadsWithRawResponse,
-    AsyncThreadsWithRawResponse,
-    ThreadsWithStreamingResponse,
-    AsyncThreadsWithStreamingResponse,
-)
 from ..._compat import cached_property
 from .chat.chat import Chat, AsyncChat
 from .assistants import (
@@ -21,7 +13,23 @@
     AsyncAssistantsWithStreamingResponse,
 )
 from ..._resource import SyncAPIResource, AsyncAPIResource
-from .vector_stores import (
+from .threads.threads import (
+    Threads,
+    AsyncThreads,
+    ThreadsWithRawResponse,
+    AsyncThreadsWithRawResponse,
+    ThreadsWithStreamingResponse,
+    AsyncThreadsWithStreamingResponse,
+)
+from .realtime.realtime import (
+    Realtime,
+    AsyncRealtime,
+    RealtimeWithRawResponse,
+    AsyncRealtimeWithRawResponse,
+    RealtimeWithStreamingResponse,
+    AsyncRealtimeWithStreamingResponse,
+)
+from .vector_stores.vector_stores import (
     VectorStores,
     AsyncVectorStores,
     VectorStoresWithRawResponse,
@@ -29,8 +37,6 @@
     VectorStoresWithStreamingResponse,
     AsyncVectorStoresWithStreamingResponse,
 )
-from .threads.threads import Threads, AsyncThreads
-from .vector_stores.vector_stores import VectorStores, AsyncVectorStores
 
 __all__ = ["Beta", "AsyncBeta"]
 
@@ -40,6 +46,10 @@ class Beta(SyncAPIResource):
     def chat(self) -> Chat:
         return Chat(self._client)
 
+    @cached_property
+    def realtime(self) -> Realtime:
+        return Realtime(self._client)
+
     @cached_property
     def vector_stores(self) -> VectorStores:
         return VectorStores(self._client)
@@ -54,10 +64,21 @@ def threads(self) -> Threads:
 
     @cached_property
     def with_raw_response(self) -> BetaWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return BetaWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> BetaWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return BetaWithStreamingResponse(self)
 
 
@@ -66,6 +87,10 @@ class AsyncBeta(AsyncAPIResource):
     def chat(self) -> AsyncChat:
         return AsyncChat(self._client)
 
+    @cached_property
+    def realtime(self) -> AsyncRealtime:
+        return AsyncRealtime(self._client)
+
     @cached_property
     def vector_stores(self) -> AsyncVectorStores:
         return AsyncVectorStores(self._client)
@@ -80,10 +105,21 @@ def threads(self) -> AsyncThreads:
 
     @cached_property
     def with_raw_response(self) -> AsyncBetaWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncBetaWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncBetaWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncBetaWithStreamingResponse(self)
 
 
@@ -91,6 +127,10 @@ class BetaWithRawResponse:
     def __init__(self, beta: Beta) -> None:
         self._beta = beta
 
+    @cached_property
+    def realtime(self) -> RealtimeWithRawResponse:
+        return RealtimeWithRawResponse(self._beta.realtime)
+
     @cached_property
     def vector_stores(self) -> VectorStoresWithRawResponse:
         return VectorStoresWithRawResponse(self._beta.vector_stores)
@@ -108,6 +148,10 @@ class AsyncBetaWithRawResponse:
     def __init__(self, beta: AsyncBeta) -> None:
         self._beta = beta
 
+    @cached_property
+    def realtime(self) -> AsyncRealtimeWithRawResponse:
+        return AsyncRealtimeWithRawResponse(self._beta.realtime)
+
     @cached_property
     def vector_stores(self) -> AsyncVectorStoresWithRawResponse:
         return AsyncVectorStoresWithRawResponse(self._beta.vector_stores)
@@ -125,6 +169,10 @@ class BetaWithStreamingResponse:
     def __init__(self, beta: Beta) -> None:
         self._beta = beta
 
+    @cached_property
+    def realtime(self) -> RealtimeWithStreamingResponse:
+        return RealtimeWithStreamingResponse(self._beta.realtime)
+
     @cached_property
     def vector_stores(self) -> VectorStoresWithStreamingResponse:
         return VectorStoresWithStreamingResponse(self._beta.vector_stores)
@@ -142,6 +190,10 @@ class AsyncBetaWithStreamingResponse:
     def __init__(self, beta: AsyncBeta) -> None:
         self._beta = beta
 
+    @cached_property
+    def realtime(self) -> AsyncRealtimeWithStreamingResponse:
+        return AsyncRealtimeWithStreamingResponse(self._beta.realtime)
+
     @cached_property
     def vector_stores(self) -> AsyncVectorStoresWithStreamingResponse:
         return AsyncVectorStoresWithStreamingResponse(self._beta.vector_stores)
diff --git a/portkey_ai/_vendor/openai/resources/beta/chat/completions.py b/portkey_ai/_vendor/openai/resources/beta/chat/completions.py
index 88ea2c05..48cb13f7 100644
--- a/portkey_ai/_vendor/openai/resources/beta/chat/completions.py
+++ b/portkey_ai/_vendor/openai/resources/beta/chat/completions.py
@@ -2,16 +2,24 @@
 
 from __future__ import annotations
 
-from typing import Dict, List, Union, Iterable, Optional
+from typing import Dict, List, Type, Union, Iterable, Optional, cast
 from functools import partial
 from typing_extensions import Literal
 
 import httpx
 
+from .... import _legacy_response
 from ...._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from ...._utils import maybe_transform, async_maybe_transform
+from ...._compat import cached_property
 from ...._resource import SyncAPIResource, AsyncAPIResource
+from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
 from ...._streaming import Stream
-from ....types.chat import completion_create_params
+from ....types.chat import (
+    ChatCompletionReasoningEffort,
+    completion_create_params,
+)
+from ...._base_client import make_request_options
 from ....lib._parsing import (
     ResponseFormatT,
     validate_input_tools as _validate_input_tools,
@@ -20,35 +28,65 @@
 )
 from ....types.chat_model import ChatModel
 from ....lib.streaming.chat import ChatCompletionStreamManager, AsyncChatCompletionStreamManager
+from ....types.chat.chat_completion import ChatCompletion
 from ....types.chat.chat_completion_chunk import ChatCompletionChunk
 from ....types.chat.parsed_chat_completion import ParsedChatCompletion
+from ....types.chat.chat_completion_modality import ChatCompletionModality
 from ....types.chat.chat_completion_tool_param import ChatCompletionToolParam
+from ....types.chat.chat_completion_audio_param import ChatCompletionAudioParam
 from ....types.chat.chat_completion_message_param import ChatCompletionMessageParam
 from ....types.chat.chat_completion_stream_options_param import ChatCompletionStreamOptionsParam
+from ....types.chat.chat_completion_prediction_content_param import ChatCompletionPredictionContentParam
 from ....types.chat.chat_completion_tool_choice_option_param import ChatCompletionToolChoiceOptionParam
 
 __all__ = ["Completions", "AsyncCompletions"]
 
 
 class Completions(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> CompletionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return CompletionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> CompletionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return CompletionsWithStreamingResponse(self)
+
     def parse(
         self,
         *,
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         response_format: type[ResponseFormatT] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
@@ -78,14 +116,17 @@ def parse(
         from pydantic import BaseModel
         from openai import OpenAI
 
+
         class Step(BaseModel):
             explanation: str
             output: str
 
+
         class MathResponse(BaseModel):
             steps: List[Step]
             final_answer: str
 
+
         client = OpenAI()
         completion = client.beta.chat.completions.parse(
             model="gpt-4o-2024-08-06",
@@ -109,38 +150,61 @@ class MathResponse(BaseModel):
             **(extra_headers or {}),
         }
 
-        raw_completion = self._client.chat.completions.create(
-            messages=messages,
-            model=model,
-            response_format=_type_to_response_format(response_format),
-            frequency_penalty=frequency_penalty,
-            function_call=function_call,
-            functions=functions,
-            logit_bias=logit_bias,
-            logprobs=logprobs,
-            max_tokens=max_tokens,
-            n=n,
-            parallel_tool_calls=parallel_tool_calls,
-            presence_penalty=presence_penalty,
-            seed=seed,
-            service_tier=service_tier,
-            stop=stop,
-            stream_options=stream_options,
-            temperature=temperature,
-            tool_choice=tool_choice,
-            tools=tools,
-            top_logprobs=top_logprobs,
-            top_p=top_p,
-            user=user,
-            extra_headers=extra_headers,
-            extra_query=extra_query,
-            extra_body=extra_body,
-            timeout=timeout,
-        )
-        return _parse_chat_completion(
-            response_format=response_format,
-            chat_completion=raw_completion,
-            input_tools=tools,
+        def parser(raw_completion: ChatCompletion) -> ParsedChatCompletion[ResponseFormatT]:
+            return _parse_chat_completion(
+                response_format=response_format,
+                chat_completion=raw_completion,
+                input_tools=tools,
+            )
+
+        return self._post(
+            "/chat/completions",
+            body=maybe_transform(
+                {
+                    "messages": messages,
+                    "model": model,
+                    "audio": audio,
+                    "frequency_penalty": frequency_penalty,
+                    "function_call": function_call,
+                    "functions": functions,
+                    "logit_bias": logit_bias,
+                    "logprobs": logprobs,
+                    "max_completion_tokens": max_completion_tokens,
+                    "max_tokens": max_tokens,
+                    "metadata": metadata,
+                    "modalities": modalities,
+                    "n": n,
+                    "parallel_tool_calls": parallel_tool_calls,
+                    "prediction": prediction,
+                    "presence_penalty": presence_penalty,
+                    "reasoning_effort": reasoning_effort,
+                    "response_format": _type_to_response_format(response_format),
+                    "seed": seed,
+                    "service_tier": service_tier,
+                    "stop": stop,
+                    "store": store,
+                    "stream": False,
+                    "stream_options": stream_options,
+                    "temperature": temperature,
+                    "tool_choice": tool_choice,
+                    "tools": tools,
+                    "top_logprobs": top_logprobs,
+                    "top_p": top_p,
+                    "user": user,
+                },
+                completion_create_params.CompletionCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                post_parser=parser,
+            ),
+            # we turn the `ChatCompletion` instance into a `ParsedChatCompletion`
+            # in the `parser` function above
+            cast_to=cast(Type[ParsedChatCompletion[ResponseFormatT]], ChatCompletion),
+            stream=False,
         )
 
     def stream(
@@ -148,19 +212,26 @@ def stream(
         *,
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         response_format: completion_create_params.ResponseFormat | type[ResponseFormatT] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
@@ -184,12 +255,12 @@ def stream(
 
         ```py
         with client.beta.chat.completions.stream(
-            model='gpt-4o-2024-08-06',
+            model="gpt-4o-2024-08-06",
             messages=[...],
         ) as stream:
             for event in stream:
-                if event.type == 'content.delta':
-                    print(event.content, flush=True, end='')
+                if event.type == "content.delta":
+                    print(event.delta, flush=True, end="")
         ```
 
         When the context manager is entered, a `ChatCompletionStream` instance is returned which, like `.create(stream=True)` is an iterator. The full list of events that are yielded by the iterator are outlined in [these docs](https://github.com/openai/openai-python/blob/main/helpers.md#chat-completions-events).
@@ -206,6 +277,7 @@ def stream(
             self._client.chat.completions.create,
             messages=messages,
             model=model,
+            audio=audio,
             stream=True,
             response_format=_type_to_response_format(response_format),
             frequency_penalty=frequency_penalty,
@@ -213,12 +285,18 @@ def stream(
             functions=functions,
             logit_bias=logit_bias,
             logprobs=logprobs,
+            max_completion_tokens=max_completion_tokens,
             max_tokens=max_tokens,
+            metadata=metadata,
+            modalities=modalities,
             n=n,
             parallel_tool_calls=parallel_tool_calls,
+            prediction=prediction,
             presence_penalty=presence_penalty,
+            reasoning_effort=reasoning_effort,
             seed=seed,
             service_tier=service_tier,
+            store=store,
             stop=stop,
             stream_options=stream_options,
             temperature=temperature,
@@ -240,24 +318,50 @@ def stream(
 
 
 class AsyncCompletions(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncCompletionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncCompletionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncCompletionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncCompletionsWithStreamingResponse(self)
+
     async def parse(
         self,
         *,
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         response_format: type[ResponseFormatT] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
@@ -287,14 +391,17 @@ async def parse(
         from pydantic import BaseModel
         from openai import AsyncOpenAI
 
+
         class Step(BaseModel):
             explanation: str
             output: str
 
+
         class MathResponse(BaseModel):
             steps: List[Step]
             final_answer: str
 
+
         client = AsyncOpenAI()
         completion = await client.beta.chat.completions.parse(
             model="gpt-4o-2024-08-06",
@@ -318,38 +425,61 @@ class MathResponse(BaseModel):
             **(extra_headers or {}),
         }
 
-        raw_completion = await self._client.chat.completions.create(
-            messages=messages,
-            model=model,
-            response_format=_type_to_response_format(response_format),
-            frequency_penalty=frequency_penalty,
-            function_call=function_call,
-            functions=functions,
-            logit_bias=logit_bias,
-            logprobs=logprobs,
-            max_tokens=max_tokens,
-            n=n,
-            parallel_tool_calls=parallel_tool_calls,
-            presence_penalty=presence_penalty,
-            seed=seed,
-            service_tier=service_tier,
-            stop=stop,
-            stream_options=stream_options,
-            temperature=temperature,
-            tool_choice=tool_choice,
-            tools=tools,
-            top_logprobs=top_logprobs,
-            top_p=top_p,
-            user=user,
-            extra_headers=extra_headers,
-            extra_query=extra_query,
-            extra_body=extra_body,
-            timeout=timeout,
-        )
-        return _parse_chat_completion(
-            response_format=response_format,
-            chat_completion=raw_completion,
-            input_tools=tools,
+        def parser(raw_completion: ChatCompletion) -> ParsedChatCompletion[ResponseFormatT]:
+            return _parse_chat_completion(
+                response_format=response_format,
+                chat_completion=raw_completion,
+                input_tools=tools,
+            )
+
+        return await self._post(
+            "/chat/completions",
+            body=await async_maybe_transform(
+                {
+                    "messages": messages,
+                    "model": model,
+                    "audio": audio,
+                    "frequency_penalty": frequency_penalty,
+                    "function_call": function_call,
+                    "functions": functions,
+                    "logit_bias": logit_bias,
+                    "logprobs": logprobs,
+                    "max_completion_tokens": max_completion_tokens,
+                    "max_tokens": max_tokens,
+                    "metadata": metadata,
+                    "modalities": modalities,
+                    "n": n,
+                    "parallel_tool_calls": parallel_tool_calls,
+                    "prediction": prediction,
+                    "presence_penalty": presence_penalty,
+                    "reasoning_effort": reasoning_effort,
+                    "response_format": _type_to_response_format(response_format),
+                    "seed": seed,
+                    "service_tier": service_tier,
+                    "store": store,
+                    "stop": stop,
+                    "stream": False,
+                    "stream_options": stream_options,
+                    "temperature": temperature,
+                    "tool_choice": tool_choice,
+                    "tools": tools,
+                    "top_logprobs": top_logprobs,
+                    "top_p": top_p,
+                    "user": user,
+                },
+                completion_create_params.CompletionCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                post_parser=parser,
+            ),
+            # we turn the `ChatCompletion` instance into a `ParsedChatCompletion`
+            # in the `parser` function above
+            cast_to=cast(Type[ParsedChatCompletion[ResponseFormatT]], ChatCompletion),
+            stream=False,
         )
 
     def stream(
@@ -357,19 +487,26 @@ def stream(
         *,
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         response_format: completion_create_params.ResponseFormat | type[ResponseFormatT] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
@@ -393,12 +530,12 @@ def stream(
 
         ```py
         async with client.beta.chat.completions.stream(
-            model='gpt-4o-2024-08-06',
+            model="gpt-4o-2024-08-06",
             messages=[...],
         ) as stream:
             async for event in stream:
-                if event.type == 'content.delta':
-                    print(event.content, flush=True, end='')
+                if event.type == "content.delta":
+                    print(event.delta, flush=True, end="")
         ```
 
         When the context manager is entered, an `AsyncChatCompletionStream` instance is returned which, like `.create(stream=True)` is an async iterator. The full list of events that are yielded by the iterator are outlined in [these docs](https://github.com/openai/openai-python/blob/main/helpers.md#chat-completions-events).
@@ -416,6 +553,7 @@ def stream(
         api_request = self._client.chat.completions.create(
             messages=messages,
             model=model,
+            audio=audio,
             stream=True,
             response_format=_type_to_response_format(response_format),
             frequency_penalty=frequency_penalty,
@@ -423,13 +561,19 @@ def stream(
             functions=functions,
             logit_bias=logit_bias,
             logprobs=logprobs,
+            max_completion_tokens=max_completion_tokens,
             max_tokens=max_tokens,
+            metadata=metadata,
+            modalities=modalities,
             n=n,
             parallel_tool_calls=parallel_tool_calls,
+            prediction=prediction,
             presence_penalty=presence_penalty,
+            reasoning_effort=reasoning_effort,
             seed=seed,
             service_tier=service_tier,
             stop=stop,
+            store=store,
             stream_options=stream_options,
             temperature=temperature,
             tool_choice=tool_choice,
@@ -447,3 +591,39 @@ def stream(
             response_format=response_format,
             input_tools=tools,
         )
+
+
+class CompletionsWithRawResponse:
+    def __init__(self, completions: Completions) -> None:
+        self._completions = completions
+
+        self.parse = _legacy_response.to_raw_response_wrapper(
+            completions.parse,
+        )
+
+
+class AsyncCompletionsWithRawResponse:
+    def __init__(self, completions: AsyncCompletions) -> None:
+        self._completions = completions
+
+        self.parse = _legacy_response.async_to_raw_response_wrapper(
+            completions.parse,
+        )
+
+
+class CompletionsWithStreamingResponse:
+    def __init__(self, completions: Completions) -> None:
+        self._completions = completions
+
+        self.parse = to_streamed_response_wrapper(
+            completions.parse,
+        )
+
+
+class AsyncCompletionsWithStreamingResponse:
+    def __init__(self, completions: AsyncCompletions) -> None:
+        self._completions = completions
+
+        self.parse = async_to_streamed_response_wrapper(
+            completions.parse,
+        )
diff --git a/portkey_ai/_vendor/openai/resources/beta/realtime/__init__.py b/portkey_ai/_vendor/openai/resources/beta/realtime/__init__.py
new file mode 100644
index 00000000..474434e6
--- /dev/null
+++ b/portkey_ai/_vendor/openai/resources/beta/realtime/__init__.py
@@ -0,0 +1,33 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .realtime import (
+    Realtime,
+    AsyncRealtime,
+    RealtimeWithRawResponse,
+    AsyncRealtimeWithRawResponse,
+    RealtimeWithStreamingResponse,
+    AsyncRealtimeWithStreamingResponse,
+)
+from .sessions import (
+    Sessions,
+    AsyncSessions,
+    SessionsWithRawResponse,
+    AsyncSessionsWithRawResponse,
+    SessionsWithStreamingResponse,
+    AsyncSessionsWithStreamingResponse,
+)
+
+__all__ = [
+    "Sessions",
+    "AsyncSessions",
+    "SessionsWithRawResponse",
+    "AsyncSessionsWithRawResponse",
+    "SessionsWithStreamingResponse",
+    "AsyncSessionsWithStreamingResponse",
+    "Realtime",
+    "AsyncRealtime",
+    "RealtimeWithRawResponse",
+    "AsyncRealtimeWithRawResponse",
+    "RealtimeWithStreamingResponse",
+    "AsyncRealtimeWithStreamingResponse",
+]
diff --git a/portkey_ai/_vendor/openai/resources/beta/realtime/realtime.py b/portkey_ai/_vendor/openai/resources/beta/realtime/realtime.py
new file mode 100644
index 00000000..c79fd462
--- /dev/null
+++ b/portkey_ai/_vendor/openai/resources/beta/realtime/realtime.py
@@ -0,0 +1,954 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+import json
+import logging
+from types import TracebackType
+from typing import TYPE_CHECKING, Any, Iterator, cast
+from typing_extensions import AsyncIterator
+
+import httpx
+from pydantic import BaseModel
+
+from .sessions import (
+    Sessions,
+    AsyncSessions,
+    SessionsWithRawResponse,
+    AsyncSessionsWithRawResponse,
+    SessionsWithStreamingResponse,
+    AsyncSessionsWithStreamingResponse,
+)
+from ...._types import NOT_GIVEN, Query, Headers, NotGiven
+from ...._utils import (
+    maybe_transform,
+    strip_not_given,
+    async_maybe_transform,
+)
+from ...._compat import cached_property
+from ...._models import construct_type_unchecked
+from ...._resource import SyncAPIResource, AsyncAPIResource
+from ...._exceptions import OpenAIError
+from ...._base_client import _merge_mappings
+from ....types.beta.realtime import session_update_event_param, response_create_event_param
+from ....types.websocket_connection_options import WebsocketConnectionOptions
+from ....types.beta.realtime.realtime_client_event import RealtimeClientEvent
+from ....types.beta.realtime.realtime_server_event import RealtimeServerEvent
+from ....types.beta.realtime.conversation_item_param import ConversationItemParam
+from ....types.beta.realtime.realtime_client_event_param import RealtimeClientEventParam
+
+if TYPE_CHECKING:
+    from websockets.sync.client import ClientConnection as WebsocketConnection
+    from websockets.asyncio.client import ClientConnection as AsyncWebsocketConnection
+
+    from ...._client import OpenAI, AsyncOpenAI
+
+__all__ = ["Realtime", "AsyncRealtime"]
+
+log: logging.Logger = logging.getLogger(__name__)
+
+
+class Realtime(SyncAPIResource):
+    @cached_property
+    def sessions(self) -> Sessions:
+        return Sessions(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> RealtimeWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return RealtimeWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> RealtimeWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return RealtimeWithStreamingResponse(self)
+
+    def connect(
+        self,
+        *,
+        model: str,
+        extra_query: Query = {},
+        extra_headers: Headers = {},
+        websocket_connection_options: WebsocketConnectionOptions = {},
+    ) -> RealtimeConnectionManager:
+        """
+        The Realtime API enables you to build low-latency, multi-modal conversational experiences. It currently supports text and audio as both input and output, as well as function calling.
+
+        Some notable benefits of the API include:
+
+        - Native speech-to-speech: Skipping an intermediate text format means low latency and nuanced output.
+        - Natural, steerable voices: The models have natural inflection and can laugh, whisper, and adhere to tone direction.
+        - Simultaneous multimodal output: Text is useful for moderation; faster-than-realtime audio ensures stable playback.
+
+        The Realtime API is a stateful, event-based API that communicates over a WebSocket.
+        """
+        return RealtimeConnectionManager(
+            client=self._client,
+            extra_query=extra_query,
+            extra_headers=extra_headers,
+            websocket_connection_options=websocket_connection_options,
+            model=model,
+        )
+
+
+class AsyncRealtime(AsyncAPIResource):
+    @cached_property
+    def sessions(self) -> AsyncSessions:
+        return AsyncSessions(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> AsyncRealtimeWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncRealtimeWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncRealtimeWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncRealtimeWithStreamingResponse(self)
+
+    def connect(
+        self,
+        *,
+        model: str,
+        extra_query: Query = {},
+        extra_headers: Headers = {},
+        websocket_connection_options: WebsocketConnectionOptions = {},
+    ) -> AsyncRealtimeConnectionManager:
+        """
+        The Realtime API enables you to build low-latency, multi-modal conversational experiences. It currently supports text and audio as both input and output, as well as function calling.
+
+        Some notable benefits of the API include:
+
+        - Native speech-to-speech: Skipping an intermediate text format means low latency and nuanced output.
+        - Natural, steerable voices: The models have natural inflection and can laugh, whisper, and adhere to tone direction.
+        - Simultaneous multimodal output: Text is useful for moderation; faster-than-realtime audio ensures stable playback.
+
+        The Realtime API is a stateful, event-based API that communicates over a WebSocket.
+        """
+        return AsyncRealtimeConnectionManager(
+            client=self._client,
+            extra_query=extra_query,
+            extra_headers=extra_headers,
+            websocket_connection_options=websocket_connection_options,
+            model=model,
+        )
+
+
+class RealtimeWithRawResponse:
+    def __init__(self, realtime: Realtime) -> None:
+        self._realtime = realtime
+
+    @cached_property
+    def sessions(self) -> SessionsWithRawResponse:
+        return SessionsWithRawResponse(self._realtime.sessions)
+
+
+class AsyncRealtimeWithRawResponse:
+    def __init__(self, realtime: AsyncRealtime) -> None:
+        self._realtime = realtime
+
+    @cached_property
+    def sessions(self) -> AsyncSessionsWithRawResponse:
+        return AsyncSessionsWithRawResponse(self._realtime.sessions)
+
+
+class RealtimeWithStreamingResponse:
+    def __init__(self, realtime: Realtime) -> None:
+        self._realtime = realtime
+
+    @cached_property
+    def sessions(self) -> SessionsWithStreamingResponse:
+        return SessionsWithStreamingResponse(self._realtime.sessions)
+
+
+class AsyncRealtimeWithStreamingResponse:
+    def __init__(self, realtime: AsyncRealtime) -> None:
+        self._realtime = realtime
+
+    @cached_property
+    def sessions(self) -> AsyncSessionsWithStreamingResponse:
+        return AsyncSessionsWithStreamingResponse(self._realtime.sessions)
+
+
+class AsyncRealtimeConnection:
+    """Represents a live websocket connection to the Realtime API"""
+
+    session: AsyncRealtimeSessionResource
+    response: AsyncRealtimeResponseResource
+    conversation: AsyncRealtimeConversationResource
+    input_audio_buffer: AsyncRealtimeInputAudioBufferResource
+
+    _connection: AsyncWebsocketConnection
+
+    def __init__(self, connection: AsyncWebsocketConnection) -> None:
+        self._connection = connection
+
+        self.session = AsyncRealtimeSessionResource(self)
+        self.response = AsyncRealtimeResponseResource(self)
+        self.conversation = AsyncRealtimeConversationResource(self)
+        self.input_audio_buffer = AsyncRealtimeInputAudioBufferResource(self)
+
+    async def __aiter__(self) -> AsyncIterator[RealtimeServerEvent]:
+        """
+        An infinite-iterator that will continue to yield events until
+        the connection is closed.
+        """
+        from websockets.exceptions import ConnectionClosedOK
+
+        try:
+            while True:
+                yield await self.recv()
+        except ConnectionClosedOK:
+            return
+
+    async def recv(self) -> RealtimeServerEvent:
+        """
+        Receive the next message from the connection and parses it into a `RealtimeServerEvent` object.
+
+        Canceling this method is safe. There's no risk of losing data.
+        """
+        return self.parse_event(await self.recv_bytes())
+
+    async def recv_bytes(self) -> bytes:
+        """Receive the next message from the connection as raw bytes.
+
+        Canceling this method is safe. There's no risk of losing data.
+
+        If you want to parse the message into a `RealtimeServerEvent` object like `.recv()` does,
+        then you can call `.parse_event(data)`.
+        """
+        message = await self._connection.recv(decode=False)
+        log.debug(f"Received websocket message: %s", message)
+        if not isinstance(message, bytes):
+            # passing `decode=False` should always result in us getting `bytes` back
+            raise TypeError(f"Expected `.recv(decode=False)` to return `bytes` but got {type(message)}")
+
+        return message
+
+    async def send(self, event: RealtimeClientEvent | RealtimeClientEventParam) -> None:
+        data = (
+            event.to_json(use_api_names=True, exclude_defaults=True, exclude_unset=True)
+            if isinstance(event, BaseModel)
+            else json.dumps(await async_maybe_transform(event, RealtimeClientEventParam))
+        )
+        await self._connection.send(data)
+
+    async def close(self, *, code: int = 1000, reason: str = "") -> None:
+        await self._connection.close(code=code, reason=reason)
+
+    def parse_event(self, data: str | bytes) -> RealtimeServerEvent:
+        """
+        Converts a raw `str` or `bytes` message into a `RealtimeServerEvent` object.
+
+        This is helpful if you're using `.recv_bytes()`.
+        """
+        return cast(
+            RealtimeServerEvent, construct_type_unchecked(value=json.loads(data), type_=cast(Any, RealtimeServerEvent))
+        )
+
+
+class AsyncRealtimeConnectionManager:
+    """
+    Context manager over a `AsyncRealtimeConnection` that is returned by `beta.realtime.connect()`
+
+    This context manager ensures that the connection will be closed when it exits.
+
+    ---
+
+    Note that if your application doesn't work well with the context manager approach then you
+    can call the `.enter()` method directly to initiate a connection.
+
+    **Warning**: You must remember to close the connection with `.close()`.
+
+    ```py
+    connection = await client.beta.realtime.connect(...).enter()
+    # ...
+    await connection.close()
+    ```
+    """
+
+    def __init__(
+        self,
+        *,
+        client: AsyncOpenAI,
+        model: str,
+        extra_query: Query,
+        extra_headers: Headers,
+        websocket_connection_options: WebsocketConnectionOptions,
+    ) -> None:
+        self.__client = client
+        self.__model = model
+        self.__connection: AsyncRealtimeConnection | None = None
+        self.__extra_query = extra_query
+        self.__extra_headers = extra_headers
+        self.__websocket_connection_options = websocket_connection_options
+
+    async def __aenter__(self) -> AsyncRealtimeConnection:
+        """
+        👋 If your application doesn't work well with the context manager approach then you
+        can call this method directly to initiate a connection.
+
+        **Warning**: You must remember to close the connection with `.close()`.
+
+        ```py
+        connection = await client.beta.realtime.connect(...).enter()
+        # ...
+        await connection.close()
+        ```
+        """
+        try:
+            from websockets.asyncio.client import connect
+        except ImportError as exc:
+            raise OpenAIError("You need to install `openai[realtime]` to use this method") from exc
+
+        url = self._prepare_url().copy_with(
+            params={
+                **self.__client.base_url.params,
+                "model": self.__model,
+                **self.__extra_query,
+            },
+        )
+        log.debug("Connecting to %s", url)
+        if self.__websocket_connection_options:
+            log.debug("Connection options: %s", self.__websocket_connection_options)
+
+        self.__connection = AsyncRealtimeConnection(
+            await connect(
+                str(url),
+                user_agent_header=self.__client.user_agent,
+                additional_headers=_merge_mappings(
+                    {
+                        **self.__client.auth_headers,
+                        "OpenAI-Beta": "realtime=v1",
+                    },
+                    self.__extra_headers,
+                ),
+                **self.__websocket_connection_options,
+            )
+        )
+
+        return self.__connection
+
+    enter = __aenter__
+
+    def _prepare_url(self) -> httpx.URL:
+        if self.__client.websocket_base_url is not None:
+            base_url = httpx.URL(self.__client.websocket_base_url)
+        else:
+            base_url = self.__client._base_url.copy_with(scheme="wss")
+
+        merge_raw_path = base_url.raw_path.rstrip(b"/") + b"/realtime"
+        return base_url.copy_with(raw_path=merge_raw_path)
+
+    async def __aexit__(
+        self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
+    ) -> None:
+        if self.__connection is not None:
+            await self.__connection.close()
+
+
+class RealtimeConnection:
+    """Represents a live websocket connection to the Realtime API"""
+
+    session: RealtimeSessionResource
+    response: RealtimeResponseResource
+    conversation: RealtimeConversationResource
+    input_audio_buffer: RealtimeInputAudioBufferResource
+
+    _connection: WebsocketConnection
+
+    def __init__(self, connection: WebsocketConnection) -> None:
+        self._connection = connection
+
+        self.session = RealtimeSessionResource(self)
+        self.response = RealtimeResponseResource(self)
+        self.conversation = RealtimeConversationResource(self)
+        self.input_audio_buffer = RealtimeInputAudioBufferResource(self)
+
+    def __iter__(self) -> Iterator[RealtimeServerEvent]:
+        """
+        An infinite-iterator that will continue to yield events until
+        the connection is closed.
+        """
+        from websockets.exceptions import ConnectionClosedOK
+
+        try:
+            while True:
+                yield self.recv()
+        except ConnectionClosedOK:
+            return
+
+    def recv(self) -> RealtimeServerEvent:
+        """
+        Receive the next message from the connection and parses it into a `RealtimeServerEvent` object.
+
+        Canceling this method is safe. There's no risk of losing data.
+        """
+        return self.parse_event(self.recv_bytes())
+
+    def recv_bytes(self) -> bytes:
+        """Receive the next message from the connection as raw bytes.
+
+        Canceling this method is safe. There's no risk of losing data.
+
+        If you want to parse the message into a `RealtimeServerEvent` object like `.recv()` does,
+        then you can call `.parse_event(data)`.
+        """
+        message = self._connection.recv(decode=False)
+        log.debug(f"Received websocket message: %s", message)
+        if not isinstance(message, bytes):
+            # passing `decode=False` should always result in us getting `bytes` back
+            raise TypeError(f"Expected `.recv(decode=False)` to return `bytes` but got {type(message)}")
+
+        return message
+
+    def send(self, event: RealtimeClientEvent | RealtimeClientEventParam) -> None:
+        data = (
+            event.to_json(use_api_names=True, exclude_defaults=True, exclude_unset=True)
+            if isinstance(event, BaseModel)
+            else json.dumps(maybe_transform(event, RealtimeClientEventParam))
+        )
+        self._connection.send(data)
+
+    def close(self, *, code: int = 1000, reason: str = "") -> None:
+        self._connection.close(code=code, reason=reason)
+
+    def parse_event(self, data: str | bytes) -> RealtimeServerEvent:
+        """
+        Converts a raw `str` or `bytes` message into a `RealtimeServerEvent` object.
+
+        This is helpful if you're using `.recv_bytes()`.
+        """
+        return cast(
+            RealtimeServerEvent, construct_type_unchecked(value=json.loads(data), type_=cast(Any, RealtimeServerEvent))
+        )
+
+
+class RealtimeConnectionManager:
+    """
+    Context manager over a `RealtimeConnection` that is returned by `beta.realtime.connect()`
+
+    This context manager ensures that the connection will be closed when it exits.
+
+    ---
+
+    Note that if your application doesn't work well with the context manager approach then you
+    can call the `.enter()` method directly to initiate a connection.
+
+    **Warning**: You must remember to close the connection with `.close()`.
+
+    ```py
+    connection = client.beta.realtime.connect(...).enter()
+    # ...
+    connection.close()
+    ```
+    """
+
+    def __init__(
+        self,
+        *,
+        client: OpenAI,
+        model: str,
+        extra_query: Query,
+        extra_headers: Headers,
+        websocket_connection_options: WebsocketConnectionOptions,
+    ) -> None:
+        self.__client = client
+        self.__model = model
+        self.__connection: RealtimeConnection | None = None
+        self.__extra_query = extra_query
+        self.__extra_headers = extra_headers
+        self.__websocket_connection_options = websocket_connection_options
+
+    def __enter__(self) -> RealtimeConnection:
+        """
+        👋 If your application doesn't work well with the context manager approach then you
+        can call this method directly to initiate a connection.
+
+        **Warning**: You must remember to close the connection with `.close()`.
+
+        ```py
+        connection = client.beta.realtime.connect(...).enter()
+        # ...
+        connection.close()
+        ```
+        """
+        try:
+            from websockets.sync.client import connect
+        except ImportError as exc:
+            raise OpenAIError("You need to install `openai[realtime]` to use this method") from exc
+
+        url = self._prepare_url().copy_with(
+            params={
+                **self.__client.base_url.params,
+                "model": self.__model,
+                **self.__extra_query,
+            },
+        )
+        log.debug("Connecting to %s", url)
+        if self.__websocket_connection_options:
+            log.debug("Connection options: %s", self.__websocket_connection_options)
+
+        self.__connection = RealtimeConnection(
+            connect(
+                str(url),
+                user_agent_header=self.__client.user_agent,
+                additional_headers=_merge_mappings(
+                    {
+                        **self.__client.auth_headers,
+                        "OpenAI-Beta": "realtime=v1",
+                    },
+                    self.__extra_headers,
+                ),
+                **self.__websocket_connection_options,
+            )
+        )
+
+        return self.__connection
+
+    enter = __enter__
+
+    def _prepare_url(self) -> httpx.URL:
+        if self.__client.websocket_base_url is not None:
+            base_url = httpx.URL(self.__client.websocket_base_url)
+        else:
+            base_url = self.__client._base_url.copy_with(scheme="wss")
+
+        merge_raw_path = base_url.raw_path.rstrip(b"/") + b"/realtime"
+        return base_url.copy_with(raw_path=merge_raw_path)
+
+    def __exit__(
+        self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
+    ) -> None:
+        if self.__connection is not None:
+            self.__connection.close()
+
+
+class BaseRealtimeConnectionResource:
+    def __init__(self, connection: RealtimeConnection) -> None:
+        self._connection = connection
+
+
+class RealtimeSessionResource(BaseRealtimeConnectionResource):
+    def update(self, *, session: session_update_event_param.Session, event_id: str | NotGiven = NOT_GIVEN) -> None:
+        """Send this event to update the session’s default configuration.
+
+        The client may
+        send this event at any time to update the session configuration, and any
+        field may be updated at any time, except for "voice". The server will respond
+        with a `session.updated` event that shows the full effective configuration.
+        Only fields that are present are updated, thus the correct way to clear a
+        field like "instructions" is to pass an empty string.
+        """
+        self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given({"type": "session.update", "session": session, "event_id": event_id}),
+            )
+        )
+
+
+class RealtimeResponseResource(BaseRealtimeConnectionResource):
+    def cancel(self, *, event_id: str | NotGiven = NOT_GIVEN, response_id: str | NotGiven = NOT_GIVEN) -> None:
+        """Send this event to cancel an in-progress response.
+
+        The server will respond
+        with a `response.cancelled` event or an error if there is no response to
+        cancel.
+        """
+        self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given({"type": "response.cancel", "event_id": event_id, "response_id": response_id}),
+            )
+        )
+
+    def create(
+        self,
+        *,
+        event_id: str | NotGiven = NOT_GIVEN,
+        response: response_create_event_param.Response | NotGiven = NOT_GIVEN,
+    ) -> None:
+        """
+        This event instructs the server to create a Response, which means triggering
+        model inference. When in Server VAD mode, the server will create Responses
+        automatically.
+
+        A Response will include at least one Item, and may have two, in which case
+        the second will be a function call. These Items will be appended to the
+        conversation history.
+
+        The server will respond with a `response.created` event, events for Items
+        and content created, and finally a `response.done` event to indicate the
+        Response is complete.
+
+        The `response.create` event includes inference configuration like
+        `instructions`, and `temperature`. These fields will override the Session's
+        configuration for this Response only.
+        """
+        self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given({"type": "response.create", "event_id": event_id, "response": response}),
+            )
+        )
+
+
+class RealtimeConversationResource(BaseRealtimeConnectionResource):
+    @cached_property
+    def item(self) -> RealtimeConversationItemResource:
+        return RealtimeConversationItemResource(self._connection)
+
+
+class RealtimeConversationItemResource(BaseRealtimeConnectionResource):
+    def delete(self, *, item_id: str, event_id: str | NotGiven = NOT_GIVEN) -> None:
+        """Send this event when you want to remove any item from the conversation
+        history.
+
+        The server will respond with a `conversation.item.deleted` event,
+        unless the item does not exist in the conversation history, in which case the
+        server will respond with an error.
+        """
+        self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given({"type": "conversation.item.delete", "item_id": item_id, "event_id": event_id}),
+            )
+        )
+
+    def create(
+        self,
+        *,
+        item: ConversationItemParam,
+        event_id: str | NotGiven = NOT_GIVEN,
+        previous_item_id: str | NotGiven = NOT_GIVEN,
+    ) -> None:
+        """
+        Add a new Item to the Conversation's context, including messages, function
+        calls, and function call responses. This event can be used both to populate a
+        "history" of the conversation and to add new items mid-stream, but has the
+        current limitation that it cannot populate assistant audio messages.
+
+        If successful, the server will respond with a `conversation.item.created`
+        event, otherwise an `error` event will be sent.
+        """
+        self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given(
+                    {
+                        "type": "conversation.item.create",
+                        "item": item,
+                        "event_id": event_id,
+                        "previous_item_id": previous_item_id,
+                    }
+                ),
+            )
+        )
+
+    def truncate(
+        self, *, audio_end_ms: int, content_index: int, item_id: str, event_id: str | NotGiven = NOT_GIVEN
+    ) -> None:
+        """Send this event to truncate a previous assistant message’s audio.
+
+        The server
+        will produce audio faster than realtime, so this event is useful when the user
+        interrupts to truncate audio that has already been sent to the client but not
+        yet played. This will synchronize the server's understanding of the audio with
+        the client's playback.
+
+        Truncating audio will delete the server-side text transcript to ensure there
+        is not text in the context that hasn't been heard by the user.
+
+        If successful, the server will respond with a `conversation.item.truncated`
+        event.
+        """
+        self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given(
+                    {
+                        "type": "conversation.item.truncate",
+                        "audio_end_ms": audio_end_ms,
+                        "content_index": content_index,
+                        "item_id": item_id,
+                        "event_id": event_id,
+                    }
+                ),
+            )
+        )
+
+
+class RealtimeInputAudioBufferResource(BaseRealtimeConnectionResource):
+    def clear(self, *, event_id: str | NotGiven = NOT_GIVEN) -> None:
+        """Send this event to clear the audio bytes in the buffer.
+
+        The server will
+        respond with an `input_audio_buffer.cleared` event.
+        """
+        self._connection.send(
+            cast(RealtimeClientEventParam, strip_not_given({"type": "input_audio_buffer.clear", "event_id": event_id}))
+        )
+
+    def commit(self, *, event_id: str | NotGiven = NOT_GIVEN) -> None:
+        """
+        Send this event to commit the user input audio buffer, which will create a
+        new user message item in the conversation. This event will produce an error
+        if the input audio buffer is empty. When in Server VAD mode, the client does
+        not need to send this event, the server will commit the audio buffer
+        automatically.
+
+        Committing the input audio buffer will trigger input audio transcription
+        (if enabled in session configuration), but it will not create a response
+        from the model. The server will respond with an `input_audio_buffer.committed`
+        event.
+        """
+        self._connection.send(
+            cast(RealtimeClientEventParam, strip_not_given({"type": "input_audio_buffer.commit", "event_id": event_id}))
+        )
+
+    def append(self, *, audio: str, event_id: str | NotGiven = NOT_GIVEN) -> None:
+        """Send this event to append audio bytes to the input audio buffer.
+
+        The audio
+        buffer is temporary storage you can write to and later commit. In Server VAD
+        mode, the audio buffer is used to detect speech and the server will decide
+        when to commit. When Server VAD is disabled, you must commit the audio buffer
+        manually.
+
+        The client may choose how much audio to place in each event up to a maximum
+        of 15 MiB, for example streaming smaller chunks from the client may allow the
+        VAD to be more responsive. Unlike made other client events, the server will
+        not send a confirmation response to this event.
+        """
+        self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given({"type": "input_audio_buffer.append", "audio": audio, "event_id": event_id}),
+            )
+        )
+
+
+class BaseAsyncRealtimeConnectionResource:
+    def __init__(self, connection: AsyncRealtimeConnection) -> None:
+        self._connection = connection
+
+
+class AsyncRealtimeSessionResource(BaseAsyncRealtimeConnectionResource):
+    async def update(
+        self, *, session: session_update_event_param.Session, event_id: str | NotGiven = NOT_GIVEN
+    ) -> None:
+        """Send this event to update the session’s default configuration.
+
+        The client may
+        send this event at any time to update the session configuration, and any
+        field may be updated at any time, except for "voice". The server will respond
+        with a `session.updated` event that shows the full effective configuration.
+        Only fields that are present are updated, thus the correct way to clear a
+        field like "instructions" is to pass an empty string.
+        """
+        await self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given({"type": "session.update", "session": session, "event_id": event_id}),
+            )
+        )
+
+
+class AsyncRealtimeResponseResource(BaseAsyncRealtimeConnectionResource):
+    async def cancel(self, *, event_id: str | NotGiven = NOT_GIVEN, response_id: str | NotGiven = NOT_GIVEN) -> None:
+        """Send this event to cancel an in-progress response.
+
+        The server will respond
+        with a `response.cancelled` event or an error if there is no response to
+        cancel.
+        """
+        await self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given({"type": "response.cancel", "event_id": event_id, "response_id": response_id}),
+            )
+        )
+
+    async def create(
+        self,
+        *,
+        event_id: str | NotGiven = NOT_GIVEN,
+        response: response_create_event_param.Response | NotGiven = NOT_GIVEN,
+    ) -> None:
+        """
+        This event instructs the server to create a Response, which means triggering
+        model inference. When in Server VAD mode, the server will create Responses
+        automatically.
+
+        A Response will include at least one Item, and may have two, in which case
+        the second will be a function call. These Items will be appended to the
+        conversation history.
+
+        The server will respond with a `response.created` event, events for Items
+        and content created, and finally a `response.done` event to indicate the
+        Response is complete.
+
+        The `response.create` event includes inference configuration like
+        `instructions`, and `temperature`. These fields will override the Session's
+        configuration for this Response only.
+        """
+        await self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given({"type": "response.create", "event_id": event_id, "response": response}),
+            )
+        )
+
+
+class AsyncRealtimeConversationResource(BaseAsyncRealtimeConnectionResource):
+    @cached_property
+    def item(self) -> AsyncRealtimeConversationItemResource:
+        return AsyncRealtimeConversationItemResource(self._connection)
+
+
+class AsyncRealtimeConversationItemResource(BaseAsyncRealtimeConnectionResource):
+    async def delete(self, *, item_id: str, event_id: str | NotGiven = NOT_GIVEN) -> None:
+        """Send this event when you want to remove any item from the conversation
+        history.
+
+        The server will respond with a `conversation.item.deleted` event,
+        unless the item does not exist in the conversation history, in which case the
+        server will respond with an error.
+        """
+        await self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given({"type": "conversation.item.delete", "item_id": item_id, "event_id": event_id}),
+            )
+        )
+
+    async def create(
+        self,
+        *,
+        item: ConversationItemParam,
+        event_id: str | NotGiven = NOT_GIVEN,
+        previous_item_id: str | NotGiven = NOT_GIVEN,
+    ) -> None:
+        """
+        Add a new Item to the Conversation's context, including messages, function
+        calls, and function call responses. This event can be used both to populate a
+        "history" of the conversation and to add new items mid-stream, but has the
+        current limitation that it cannot populate assistant audio messages.
+
+        If successful, the server will respond with a `conversation.item.created`
+        event, otherwise an `error` event will be sent.
+        """
+        await self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given(
+                    {
+                        "type": "conversation.item.create",
+                        "item": item,
+                        "event_id": event_id,
+                        "previous_item_id": previous_item_id,
+                    }
+                ),
+            )
+        )
+
+    async def truncate(
+        self, *, audio_end_ms: int, content_index: int, item_id: str, event_id: str | NotGiven = NOT_GIVEN
+    ) -> None:
+        """Send this event to truncate a previous assistant message’s audio.
+
+        The server
+        will produce audio faster than realtime, so this event is useful when the user
+        interrupts to truncate audio that has already been sent to the client but not
+        yet played. This will synchronize the server's understanding of the audio with
+        the client's playback.
+
+        Truncating audio will delete the server-side text transcript to ensure there
+        is not text in the context that hasn't been heard by the user.
+
+        If successful, the server will respond with a `conversation.item.truncated`
+        event.
+        """
+        await self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given(
+                    {
+                        "type": "conversation.item.truncate",
+                        "audio_end_ms": audio_end_ms,
+                        "content_index": content_index,
+                        "item_id": item_id,
+                        "event_id": event_id,
+                    }
+                ),
+            )
+        )
+
+
+class AsyncRealtimeInputAudioBufferResource(BaseAsyncRealtimeConnectionResource):
+    async def clear(self, *, event_id: str | NotGiven = NOT_GIVEN) -> None:
+        """Send this event to clear the audio bytes in the buffer.
+
+        The server will
+        respond with an `input_audio_buffer.cleared` event.
+        """
+        await self._connection.send(
+            cast(RealtimeClientEventParam, strip_not_given({"type": "input_audio_buffer.clear", "event_id": event_id}))
+        )
+
+    async def commit(self, *, event_id: str | NotGiven = NOT_GIVEN) -> None:
+        """
+        Send this event to commit the user input audio buffer, which will create a
+        new user message item in the conversation. This event will produce an error
+        if the input audio buffer is empty. When in Server VAD mode, the client does
+        not need to send this event, the server will commit the audio buffer
+        automatically.
+
+        Committing the input audio buffer will trigger input audio transcription
+        (if enabled in session configuration), but it will not create a response
+        from the model. The server will respond with an `input_audio_buffer.committed`
+        event.
+        """
+        await self._connection.send(
+            cast(RealtimeClientEventParam, strip_not_given({"type": "input_audio_buffer.commit", "event_id": event_id}))
+        )
+
+    async def append(self, *, audio: str, event_id: str | NotGiven = NOT_GIVEN) -> None:
+        """Send this event to append audio bytes to the input audio buffer.
+
+        The audio
+        buffer is temporary storage you can write to and later commit. In Server VAD
+        mode, the audio buffer is used to detect speech and the server will decide
+        when to commit. When Server VAD is disabled, you must commit the audio buffer
+        manually.
+
+        The client may choose how much audio to place in each event up to a maximum
+        of 15 MiB, for example streaming smaller chunks from the client may allow the
+        VAD to be more responsive. Unlike made other client events, the server will
+        not send a confirmation response to this event.
+        """
+        await self._connection.send(
+            cast(
+                RealtimeClientEventParam,
+                strip_not_given({"type": "input_audio_buffer.append", "audio": audio, "event_id": event_id}),
+            )
+        )
diff --git a/portkey_ai/_vendor/openai/resources/beta/realtime/sessions.py b/portkey_ai/_vendor/openai/resources/beta/realtime/sessions.py
new file mode 100644
index 00000000..1d1ee701
--- /dev/null
+++ b/portkey_ai/_vendor/openai/resources/beta/realtime/sessions.py
@@ -0,0 +1,337 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import List, Union, Iterable
+from typing_extensions import Literal
+
+import httpx
+
+from .... import _legacy_response
+from ...._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from ...._utils import (
+    maybe_transform,
+    async_maybe_transform,
+)
+from ...._compat import cached_property
+from ...._resource import SyncAPIResource, AsyncAPIResource
+from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ...._base_client import make_request_options
+from ....types.beta.realtime import session_create_params
+from ....types.beta.realtime.session_create_response import SessionCreateResponse
+
+__all__ = ["Sessions", "AsyncSessions"]
+
+
+class Sessions(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> SessionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return SessionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> SessionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return SessionsWithStreamingResponse(self)
+
+    def create(
+        self,
+        *,
+        model: Literal[
+            "gpt-4o-realtime-preview",
+            "gpt-4o-realtime-preview-2024-10-01",
+            "gpt-4o-realtime-preview-2024-12-17",
+            "gpt-4o-mini-realtime-preview",
+            "gpt-4o-mini-realtime-preview-2024-12-17",
+        ],
+        input_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        input_audio_transcription: session_create_params.InputAudioTranscription | NotGiven = NOT_GIVEN,
+        instructions: str | NotGiven = NOT_GIVEN,
+        max_response_output_tokens: Union[int, Literal["inf"]] | NotGiven = NOT_GIVEN,
+        modalities: List[Literal["text", "audio"]] | NotGiven = NOT_GIVEN,
+        output_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        tool_choice: str | NotGiven = NOT_GIVEN,
+        tools: Iterable[session_create_params.Tool] | NotGiven = NOT_GIVEN,
+        turn_detection: session_create_params.TurnDetection | NotGiven = NOT_GIVEN,
+        voice: Literal["alloy", "ash", "ballad", "coral", "echo", "sage", "shimmer", "verse"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SessionCreateResponse:
+        """
+        Create an ephemeral API token for use in client-side applications with the
+        Realtime API. Can be configured with the same session parameters as the
+        `session.update` client event.
+
+        It responds with a session object, plus a `client_secret` key which contains a
+        usable ephemeral API token that can be used to authenticate browser clients for
+        the Realtime API.
+
+        Args:
+          model: The Realtime model used for this session.
+
+          input_audio_format: The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
+
+          input_audio_transcription: Configuration for input audio transcription, defaults to off and can be set to
+              `null` to turn off once on. Input audio transcription is not native to the
+              model, since the model consumes audio directly. Transcription runs
+              asynchronously through Whisper and should be treated as rough guidance rather
+              than the representation understood by the model.
+
+          instructions: The default system instructions (i.e. system message) prepended to model calls.
+              This field allows the client to guide the model on desired responses. The model
+              can be instructed on response content and format, (e.g. "be extremely succinct",
+              "act friendly", "here are examples of good responses") and on audio behavior
+              (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The
+              instructions are not guaranteed to be followed by the model, but they provide
+              guidance to the model on the desired behavior.
+
+              Note that the server sets default instructions which will be used if this field
+              is not set and are visible in the `session.created` event at the start of the
+              session.
+
+          max_response_output_tokens: Maximum number of output tokens for a single assistant response, inclusive of
+              tool calls. Provide an integer between 1 and 4096 to limit output tokens, or
+              `inf` for the maximum available tokens for a given model. Defaults to `inf`.
+
+          modalities: The set of modalities the model can respond with. To disable audio, set this to
+              ["text"].
+
+          output_audio_format: The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
+
+          temperature: Sampling temperature for the model, limited to [0.6, 1.2]. Defaults to 0.8.
+
+          tool_choice: How the model chooses tools. Options are `auto`, `none`, `required`, or specify
+              a function.
+
+          tools: Tools (functions) available to the model.
+
+          turn_detection: Configuration for turn detection. Can be set to `null` to turn off. Server VAD
+              means that the model will detect the start and end of speech based on audio
+              volume and respond at the end of user speech.
+
+          voice: The voice the model uses to respond. Voice cannot be changed during the session
+              once the model has responded with audio at least once. Current voice options are
+              `alloy`, `ash`, `ballad`, `coral`, `echo` `sage`, `shimmer` and `verse`.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._post(
+            "/realtime/sessions",
+            body=maybe_transform(
+                {
+                    "model": model,
+                    "input_audio_format": input_audio_format,
+                    "input_audio_transcription": input_audio_transcription,
+                    "instructions": instructions,
+                    "max_response_output_tokens": max_response_output_tokens,
+                    "modalities": modalities,
+                    "output_audio_format": output_audio_format,
+                    "temperature": temperature,
+                    "tool_choice": tool_choice,
+                    "tools": tools,
+                    "turn_detection": turn_detection,
+                    "voice": voice,
+                },
+                session_create_params.SessionCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=SessionCreateResponse,
+        )
+
+
+class AsyncSessions(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncSessionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncSessionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncSessionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncSessionsWithStreamingResponse(self)
+
+    async def create(
+        self,
+        *,
+        model: Literal[
+            "gpt-4o-realtime-preview",
+            "gpt-4o-realtime-preview-2024-10-01",
+            "gpt-4o-realtime-preview-2024-12-17",
+            "gpt-4o-mini-realtime-preview",
+            "gpt-4o-mini-realtime-preview-2024-12-17",
+        ],
+        input_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        input_audio_transcription: session_create_params.InputAudioTranscription | NotGiven = NOT_GIVEN,
+        instructions: str | NotGiven = NOT_GIVEN,
+        max_response_output_tokens: Union[int, Literal["inf"]] | NotGiven = NOT_GIVEN,
+        modalities: List[Literal["text", "audio"]] | NotGiven = NOT_GIVEN,
+        output_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        tool_choice: str | NotGiven = NOT_GIVEN,
+        tools: Iterable[session_create_params.Tool] | NotGiven = NOT_GIVEN,
+        turn_detection: session_create_params.TurnDetection | NotGiven = NOT_GIVEN,
+        voice: Literal["alloy", "ash", "ballad", "coral", "echo", "sage", "shimmer", "verse"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SessionCreateResponse:
+        """
+        Create an ephemeral API token for use in client-side applications with the
+        Realtime API. Can be configured with the same session parameters as the
+        `session.update` client event.
+
+        It responds with a session object, plus a `client_secret` key which contains a
+        usable ephemeral API token that can be used to authenticate browser clients for
+        the Realtime API.
+
+        Args:
+          model: The Realtime model used for this session.
+
+          input_audio_format: The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
+
+          input_audio_transcription: Configuration for input audio transcription, defaults to off and can be set to
+              `null` to turn off once on. Input audio transcription is not native to the
+              model, since the model consumes audio directly. Transcription runs
+              asynchronously through Whisper and should be treated as rough guidance rather
+              than the representation understood by the model.
+
+          instructions: The default system instructions (i.e. system message) prepended to model calls.
+              This field allows the client to guide the model on desired responses. The model
+              can be instructed on response content and format, (e.g. "be extremely succinct",
+              "act friendly", "here are examples of good responses") and on audio behavior
+              (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The
+              instructions are not guaranteed to be followed by the model, but they provide
+              guidance to the model on the desired behavior.
+
+              Note that the server sets default instructions which will be used if this field
+              is not set and are visible in the `session.created` event at the start of the
+              session.
+
+          max_response_output_tokens: Maximum number of output tokens for a single assistant response, inclusive of
+              tool calls. Provide an integer between 1 and 4096 to limit output tokens, or
+              `inf` for the maximum available tokens for a given model. Defaults to `inf`.
+
+          modalities: The set of modalities the model can respond with. To disable audio, set this to
+              ["text"].
+
+          output_audio_format: The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
+
+          temperature: Sampling temperature for the model, limited to [0.6, 1.2]. Defaults to 0.8.
+
+          tool_choice: How the model chooses tools. Options are `auto`, `none`, `required`, or specify
+              a function.
+
+          tools: Tools (functions) available to the model.
+
+          turn_detection: Configuration for turn detection. Can be set to `null` to turn off. Server VAD
+              means that the model will detect the start and end of speech based on audio
+              volume and respond at the end of user speech.
+
+          voice: The voice the model uses to respond. Voice cannot be changed during the session
+              once the model has responded with audio at least once. Current voice options are
+              `alloy`, `ash`, `ballad`, `coral`, `echo` `sage`, `shimmer` and `verse`.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._post(
+            "/realtime/sessions",
+            body=await async_maybe_transform(
+                {
+                    "model": model,
+                    "input_audio_format": input_audio_format,
+                    "input_audio_transcription": input_audio_transcription,
+                    "instructions": instructions,
+                    "max_response_output_tokens": max_response_output_tokens,
+                    "modalities": modalities,
+                    "output_audio_format": output_audio_format,
+                    "temperature": temperature,
+                    "tool_choice": tool_choice,
+                    "tools": tools,
+                    "turn_detection": turn_detection,
+                    "voice": voice,
+                },
+                session_create_params.SessionCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=SessionCreateResponse,
+        )
+
+
+class SessionsWithRawResponse:
+    def __init__(self, sessions: Sessions) -> None:
+        self._sessions = sessions
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            sessions.create,
+        )
+
+
+class AsyncSessionsWithRawResponse:
+    def __init__(self, sessions: AsyncSessions) -> None:
+        self._sessions = sessions
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            sessions.create,
+        )
+
+
+class SessionsWithStreamingResponse:
+    def __init__(self, sessions: Sessions) -> None:
+        self._sessions = sessions
+
+        self.create = to_streamed_response_wrapper(
+            sessions.create,
+        )
+
+
+class AsyncSessionsWithStreamingResponse:
+    def __init__(self, sessions: AsyncSessions) -> None:
+        self._sessions = sessions
+
+        self.create = async_to_streamed_response_wrapper(
+            sessions.create,
+        )
diff --git a/portkey_ai/_vendor/openai/resources/beta/threads/messages.py b/portkey_ai/_vendor/openai/resources/beta/threads/messages.py
index f0832515..e8485073 100644
--- a/portkey_ai/_vendor/openai/resources/beta/threads/messages.py
+++ b/portkey_ai/_vendor/openai/resources/beta/threads/messages.py
@@ -32,10 +32,21 @@
 class Messages(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> MessagesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return MessagesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> MessagesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return MessagesWithStreamingResponse(self)
 
     def create(
@@ -71,7 +82,7 @@ def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           extra_headers: Send extra headers
@@ -158,7 +169,7 @@ def update(
         Args:
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           extra_headers: Send extra headers
@@ -210,8 +221,8 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           limit: A limit on the number of objects to be returned. Limit can range between 1 and
               100, and the default is 20.
@@ -295,10 +306,21 @@ def delete(
 class AsyncMessages(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncMessagesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncMessagesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncMessagesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncMessagesWithStreamingResponse(self)
 
     async def create(
@@ -334,7 +356,7 @@ async def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           extra_headers: Send extra headers
@@ -421,7 +443,7 @@ async def update(
         Args:
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           extra_headers: Send extra headers
@@ -473,8 +495,8 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           limit: A limit on the number of objects to be returned. Limit can range between 1 and
               100, and the default is 20.
diff --git a/portkey_ai/_vendor/openai/resources/beta/threads/runs/runs.py b/portkey_ai/_vendor/openai/resources/beta/threads/runs/runs.py
index cbfb9546..0418d570 100644
--- a/portkey_ai/_vendor/openai/resources/beta/threads/runs/runs.py
+++ b/portkey_ai/_vendor/openai/resources/beta/threads/runs/runs.py
@@ -3,9 +3,9 @@
 from __future__ import annotations
 
 import typing_extensions
-from typing import Union, Iterable, Optional, overload
+from typing import List, Union, Iterable, Optional
 from functools import partial
-from typing_extensions import Literal
+from typing_extensions import Literal, overload
 
 import httpx
 
@@ -49,6 +49,7 @@
 from .....types.beta.threads.run import Run
 from .....types.beta.assistant_tool_param import AssistantToolParam
 from .....types.beta.assistant_stream_event import AssistantStreamEvent
+from .....types.beta.threads.runs.run_step_include import RunStepInclude
 from .....types.beta.assistant_tool_choice_option_param import AssistantToolChoiceOptionParam
 from .....types.beta.assistant_response_format_option_param import AssistantResponseFormatOptionParam
 
@@ -62,10 +63,21 @@ def steps(self) -> Steps:
 
     @cached_property
     def with_raw_response(self) -> RunsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return RunsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> RunsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return RunsWithStreamingResponse(self)
 
     @overload
@@ -74,6 +86,7 @@ def create(
         thread_id: str,
         *,
         assistant_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -104,6 +117,14 @@ def create(
               [assistant](https://platform.openai.com/docs/api-reference/assistants) to use to
               execute this run.
 
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+              for more information.
+
           additional_instructions: Appends additional instructions at the end of the instructions for the run. This
               is useful for modifying the behavior on a per-run basis without overriding other
               instructions.
@@ -128,7 +149,7 @@ def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -137,20 +158,20 @@ def create(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -206,6 +227,7 @@ def create(
         *,
         assistant_id: str,
         stream: Literal[True],
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -239,6 +261,14 @@ def create(
               events, terminating when the Run enters a terminal state with a `data: [DONE]`
               message.
 
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+              for more information.
+
           additional_instructions: Appends additional instructions at the end of the instructions for the run. This
               is useful for modifying the behavior on a per-run basis without overriding other
               instructions.
@@ -263,7 +293,7 @@ def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -272,20 +302,20 @@ def create(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -337,6 +367,7 @@ def create(
         *,
         assistant_id: str,
         stream: bool,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -370,6 +401,14 @@ def create(
               events, terminating when the Run enters a terminal state with a `data: [DONE]`
               message.
 
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+              for more information.
+
           additional_instructions: Appends additional instructions at the end of the instructions for the run. This
               is useful for modifying the behavior on a per-run basis without overriding other
               instructions.
@@ -394,7 +433,7 @@ def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -403,20 +442,20 @@ def create(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -467,6 +506,7 @@ def create(
         thread_id: str,
         *,
         assistant_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -516,7 +556,11 @@ def create(
                 run_create_params.RunCreateParams,
             ),
             options=make_request_options(
-                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform({"include": include}, run_create_params.RunCreateParams),
             ),
             cast_to=Run,
             stream=stream or False,
@@ -579,7 +623,7 @@ def update(
         Args:
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           extra_headers: Send extra headers
@@ -630,8 +674,8 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           limit: A limit on the number of objects to be returned. Limit can range between 1 and
               100, and the default is 20.
@@ -712,6 +756,7 @@ def create_and_poll(
         self,
         *,
         assistant_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -743,6 +788,7 @@ def create_and_poll(
         run = self.create(
             thread_id=thread_id,
             assistant_id=assistant_id,
+            include=include,
             additional_instructions=additional_instructions,
             additional_messages=additional_messages,
             instructions=instructions,
@@ -958,6 +1004,7 @@ def stream(
         self,
         *,
         assistant_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -988,6 +1035,7 @@ def stream(
         self,
         *,
         assistant_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -1018,6 +1066,7 @@ def stream(
         self,
         *,
         assistant_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -1076,7 +1125,11 @@ def stream(
                 run_create_params.RunCreateParams,
             ),
             options=make_request_options(
-                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform({"include": include}, run_create_params.RunCreateParams),
             ),
             cast_to=Run,
             stream=True,
@@ -1375,10 +1428,21 @@ def steps(self) -> AsyncSteps:
 
     @cached_property
     def with_raw_response(self) -> AsyncRunsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncRunsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncRunsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncRunsWithStreamingResponse(self)
 
     @overload
@@ -1387,6 +1451,7 @@ async def create(
         thread_id: str,
         *,
         assistant_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -1417,6 +1482,14 @@ async def create(
               [assistant](https://platform.openai.com/docs/api-reference/assistants) to use to
               execute this run.
 
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+              for more information.
+
           additional_instructions: Appends additional instructions at the end of the instructions for the run. This
               is useful for modifying the behavior on a per-run basis without overriding other
               instructions.
@@ -1441,7 +1514,7 @@ async def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -1450,20 +1523,20 @@ async def create(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -1519,6 +1592,7 @@ async def create(
         *,
         assistant_id: str,
         stream: Literal[True],
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -1552,6 +1626,14 @@ async def create(
               events, terminating when the Run enters a terminal state with a `data: [DONE]`
               message.
 
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+              for more information.
+
           additional_instructions: Appends additional instructions at the end of the instructions for the run. This
               is useful for modifying the behavior on a per-run basis without overriding other
               instructions.
@@ -1576,7 +1658,7 @@ async def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -1585,20 +1667,20 @@ async def create(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -1650,6 +1732,7 @@ async def create(
         *,
         assistant_id: str,
         stream: bool,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -1683,6 +1766,14 @@ async def create(
               events, terminating when the Run enters a terminal state with a `data: [DONE]`
               message.
 
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+              for more information.
+
           additional_instructions: Appends additional instructions at the end of the instructions for the run. This
               is useful for modifying the behavior on a per-run basis without overriding other
               instructions.
@@ -1707,7 +1798,7 @@ async def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -1716,20 +1807,20 @@ async def create(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -1780,6 +1871,7 @@ async def create(
         thread_id: str,
         *,
         assistant_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -1829,7 +1921,11 @@ async def create(
                 run_create_params.RunCreateParams,
             ),
             options=make_request_options(
-                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=await async_maybe_transform({"include": include}, run_create_params.RunCreateParams),
             ),
             cast_to=Run,
             stream=stream or False,
@@ -1892,7 +1988,7 @@ async def update(
         Args:
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           extra_headers: Send extra headers
@@ -1943,8 +2039,8 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           limit: A limit on the number of objects to be returned. Limit can range between 1 and
               100, and the default is 20.
@@ -2025,6 +2121,7 @@ async def create_and_poll(
         self,
         *,
         assistant_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -2056,6 +2153,7 @@ async def create_and_poll(
         run = await self.create(
             thread_id=thread_id,
             assistant_id=assistant_id,
+            include=include,
             additional_instructions=additional_instructions,
             additional_messages=additional_messages,
             instructions=instructions,
@@ -2303,6 +2401,7 @@ def stream(
         self,
         *,
         assistant_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -2333,6 +2432,7 @@ def stream(
         self,
         *,
         assistant_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         additional_instructions: Optional[str] | NotGiven = NOT_GIVEN,
         additional_messages: Optional[Iterable[run_create_params.AdditionalMessage]] | NotGiven = NOT_GIVEN,
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
@@ -2393,7 +2493,11 @@ def stream(
                 run_create_params.RunCreateParams,
             ),
             options=make_request_options(
-                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform({"include": include}, run_create_params.RunCreateParams),
             ),
             cast_to=Run,
             stream=True,
diff --git a/portkey_ai/_vendor/openai/resources/beta/threads/runs/steps.py b/portkey_ai/_vendor/openai/resources/beta/threads/runs/steps.py
index 51200893..9bd91e39 100644
--- a/portkey_ai/_vendor/openai/resources/beta/threads/runs/steps.py
+++ b/portkey_ai/_vendor/openai/resources/beta/threads/runs/steps.py
@@ -2,23 +2,25 @@
 
 from __future__ import annotations
 
+from typing import List
 from typing_extensions import Literal
 
 import httpx
 
 from ..... import _legacy_response
 from ....._types import NOT_GIVEN, Body, Query, Headers, NotGiven
-from ....._utils import maybe_transform
+from ....._utils import (
+    maybe_transform,
+    async_maybe_transform,
+)
 from ....._compat import cached_property
 from ....._resource import SyncAPIResource, AsyncAPIResource
 from ....._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
 from .....pagination import SyncCursorPage, AsyncCursorPage
-from ....._base_client import (
-    AsyncPaginator,
-    make_request_options,
-)
-from .....types.beta.threads.runs import step_list_params
+from ....._base_client import AsyncPaginator, make_request_options
+from .....types.beta.threads.runs import step_list_params, step_retrieve_params
 from .....types.beta.threads.runs.run_step import RunStep
+from .....types.beta.threads.runs.run_step_include import RunStepInclude
 
 __all__ = ["Steps", "AsyncSteps"]
 
@@ -26,10 +28,21 @@
 class Steps(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> StepsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return StepsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> StepsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return StepsWithStreamingResponse(self)
 
     def retrieve(
@@ -38,6 +51,7 @@ def retrieve(
         *,
         thread_id: str,
         run_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -49,6 +63,14 @@ def retrieve(
         Retrieves a run step.
 
         Args:
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+              for more information.
+
           extra_headers: Send extra headers
 
           extra_query: Add additional query parameters to the request
@@ -67,7 +89,11 @@ def retrieve(
         return self._get(
             f"/threads/{thread_id}/runs/{run_id}/steps/{step_id}",
             options=make_request_options(
-                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform({"include": include}, step_retrieve_params.StepRetrieveParams),
             ),
             cast_to=RunStep,
         )
@@ -79,6 +105,7 @@ def list(
         thread_id: str,
         after: str | NotGiven = NOT_GIVEN,
         before: str | NotGiven = NOT_GIVEN,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         limit: int | NotGiven = NOT_GIVEN,
         order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -99,8 +126,16 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
+
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+              for more information.
 
           limit: A limit on the number of objects to be returned. Limit can range between 1 and
               100, and the default is 20.
@@ -133,6 +168,7 @@ def list(
                     {
                         "after": after,
                         "before": before,
+                        "include": include,
                         "limit": limit,
                         "order": order,
                     },
@@ -146,10 +182,21 @@ def list(
 class AsyncSteps(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncStepsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncStepsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncStepsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncStepsWithStreamingResponse(self)
 
     async def retrieve(
@@ -158,6 +205,7 @@ async def retrieve(
         *,
         thread_id: str,
         run_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -169,6 +217,14 @@ async def retrieve(
         Retrieves a run step.
 
         Args:
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+              for more information.
+
           extra_headers: Send extra headers
 
           extra_query: Add additional query parameters to the request
@@ -187,7 +243,11 @@ async def retrieve(
         return await self._get(
             f"/threads/{thread_id}/runs/{run_id}/steps/{step_id}",
             options=make_request_options(
-                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=await async_maybe_transform({"include": include}, step_retrieve_params.StepRetrieveParams),
             ),
             cast_to=RunStep,
         )
@@ -199,6 +259,7 @@ def list(
         thread_id: str,
         after: str | NotGiven = NOT_GIVEN,
         before: str | NotGiven = NOT_GIVEN,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
         limit: int | NotGiven = NOT_GIVEN,
         order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -219,8 +280,16 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
+
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+              for more information.
 
           limit: A limit on the number of objects to be returned. Limit can range between 1 and
               100, and the default is 20.
@@ -253,6 +322,7 @@ def list(
                     {
                         "after": after,
                         "before": before,
+                        "include": include,
                         "limit": limit,
                         "order": order,
                     },
diff --git a/portkey_ai/_vendor/openai/resources/beta/threads/threads.py b/portkey_ai/_vendor/openai/resources/beta/threads/threads.py
index 4c95c484..e45090ab 100644
--- a/portkey_ai/_vendor/openai/resources/beta/threads/threads.py
+++ b/portkey_ai/_vendor/openai/resources/beta/threads/threads.py
@@ -2,21 +2,13 @@
 
 from __future__ import annotations
 
-from typing import Union, Iterable, Optional, overload
+from typing import Union, Iterable, Optional
 from functools import partial
-from typing_extensions import Literal
+from typing_extensions import Literal, overload
 
 import httpx
 
 from .... import _legacy_response
-from .runs import (
-    Runs,
-    AsyncRuns,
-    RunsWithRawResponse,
-    AsyncRunsWithRawResponse,
-    RunsWithStreamingResponse,
-    AsyncRunsWithStreamingResponse,
-)
 from .messages import (
     Messages,
     AsyncMessages,
@@ -31,7 +23,14 @@
     maybe_transform,
     async_maybe_transform,
 )
-from .runs.runs import Runs, AsyncRuns
+from .runs.runs import (
+    Runs,
+    AsyncRuns,
+    RunsWithRawResponse,
+    AsyncRunsWithRawResponse,
+    RunsWithStreamingResponse,
+    AsyncRunsWithStreamingResponse,
+)
 from ...._compat import cached_property
 from ...._resource import SyncAPIResource, AsyncAPIResource
 from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
@@ -72,10 +71,21 @@ def messages(self) -> Messages:
 
     @cached_property
     def with_raw_response(self) -> ThreadsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return ThreadsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> ThreadsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return ThreadsWithStreamingResponse(self)
 
     def create(
@@ -100,7 +110,7 @@ def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           tool_resources: A set of resources that are made available to the assistant's tools in this
@@ -186,7 +196,7 @@ def update(
         Args:
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           tool_resources: A set of resources that are made available to the assistant's tools in this
@@ -306,7 +316,7 @@ def create_and_run(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -315,20 +325,20 @@ def create_and_run(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -440,7 +450,7 @@ def create_and_run(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -449,20 +459,20 @@ def create_and_run(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -570,7 +580,7 @@ def create_and_run(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -579,20 +589,20 @@ def create_and_run(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -895,10 +905,21 @@ def messages(self) -> AsyncMessages:
 
     @cached_property
     def with_raw_response(self) -> AsyncThreadsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncThreadsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncThreadsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncThreadsWithStreamingResponse(self)
 
     async def create(
@@ -923,7 +944,7 @@ async def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           tool_resources: A set of resources that are made available to the assistant's tools in this
@@ -1009,7 +1030,7 @@ async def update(
         Args:
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           tool_resources: A set of resources that are made available to the assistant's tools in this
@@ -1129,7 +1150,7 @@ async def create_and_run(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -1138,20 +1159,20 @@ async def create_and_run(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -1263,7 +1284,7 @@ async def create_and_run(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -1272,20 +1293,20 @@ async def create_and_run(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -1393,7 +1414,7 @@ async def create_and_run(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           model: The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to
@@ -1402,20 +1423,20 @@ async def create_and_run(
               assistant will be used.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
           response_format: Specifies the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
               and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
               Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-              Outputs which guarantees the model will match your supplied JSON schema. Learn
-              more in the
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
               [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
diff --git a/portkey_ai/_vendor/openai/resources/beta/vector_stores/file_batches.py b/portkey_ai/_vendor/openai/resources/beta/vector_stores/file_batches.py
index d6862c24..9f9e643b 100644
--- a/portkey_ai/_vendor/openai/resources/beta/vector_stores/file_batches.py
+++ b/portkey_ai/_vendor/openai/resources/beta/vector_stores/file_batches.py
@@ -22,11 +22,10 @@
 from ...._resource import SyncAPIResource, AsyncAPIResource
 from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
 from ....pagination import SyncCursorPage, AsyncCursorPage
-from ...._base_client import (
-    AsyncPaginator,
-    make_request_options,
-)
+from ....types.beta import FileChunkingStrategyParam
+from ...._base_client import AsyncPaginator, make_request_options
 from ....types.beta.vector_stores import file_batch_create_params, file_batch_list_files_params
+from ....types.beta.file_chunking_strategy_param import FileChunkingStrategyParam
 from ....types.beta.vector_stores.vector_store_file import VectorStoreFile
 from ....types.beta.vector_stores.vector_store_file_batch import VectorStoreFileBatch
 
@@ -36,10 +35,21 @@
 class FileBatches(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> FileBatchesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return FileBatchesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> FileBatchesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return FileBatchesWithStreamingResponse(self)
 
     def create(
@@ -47,7 +57,7 @@ def create(
         vector_store_id: str,
         *,
         file_ids: List[str],
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -64,7 +74,7 @@ def create(
               files.
 
           chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
-              strategy.
+              strategy. Only applicable if `file_ids` is non-empty.
 
           extra_headers: Send extra headers
 
@@ -174,7 +184,7 @@ def create_and_poll(
         *,
         file_ids: List[str],
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFileBatch:
         """Create a vector store batch and poll until all files have been processed."""
         batch = self.create(
@@ -217,8 +227,8 @@ def list_files(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`.
 
@@ -308,7 +318,7 @@ def upload_and_poll(
         max_concurrency: int = 5,
         file_ids: List[str] = [],
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFileBatch:
         """Uploads the given files concurrently and then creates a vector store file batch.
 
@@ -354,10 +364,21 @@ def upload_and_poll(
 class AsyncFileBatches(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncFileBatchesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncFileBatchesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncFileBatchesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncFileBatchesWithStreamingResponse(self)
 
     async def create(
@@ -365,7 +386,7 @@ async def create(
         vector_store_id: str,
         *,
         file_ids: List[str],
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -382,7 +403,7 @@ async def create(
               files.
 
           chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
-              strategy.
+              strategy. Only applicable if `file_ids` is non-empty.
 
           extra_headers: Send extra headers
 
@@ -492,7 +513,7 @@ async def create_and_poll(
         *,
         file_ids: List[str],
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFileBatch:
         """Create a vector store batch and poll until all files have been processed."""
         batch = await self.create(
@@ -535,8 +556,8 @@ def list_files(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`.
 
@@ -626,7 +647,7 @@ async def upload_and_poll(
         max_concurrency: int = 5,
         file_ids: List[str] = [],
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_batch_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFileBatch:
         """Uploads the given files concurrently and then creates a vector store file batch.
 
diff --git a/portkey_ai/_vendor/openai/resources/beta/vector_stores/files.py b/portkey_ai/_vendor/openai/resources/beta/vector_stores/files.py
index 35ca331c..7c155ac9 100644
--- a/portkey_ai/_vendor/openai/resources/beta/vector_stores/files.py
+++ b/portkey_ai/_vendor/openai/resources/beta/vector_stores/files.py
@@ -18,11 +18,10 @@
 from ...._resource import SyncAPIResource, AsyncAPIResource
 from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
 from ....pagination import SyncCursorPage, AsyncCursorPage
-from ...._base_client import (
-    AsyncPaginator,
-    make_request_options,
-)
+from ....types.beta import FileChunkingStrategyParam
+from ...._base_client import AsyncPaginator, make_request_options
 from ....types.beta.vector_stores import file_list_params, file_create_params
+from ....types.beta.file_chunking_strategy_param import FileChunkingStrategyParam
 from ....types.beta.vector_stores.vector_store_file import VectorStoreFile
 from ....types.beta.vector_stores.vector_store_file_deleted import VectorStoreFileDeleted
 
@@ -32,10 +31,21 @@
 class Files(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> FilesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return FilesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> FilesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return FilesWithStreamingResponse(self)
 
     def create(
@@ -43,7 +53,7 @@ def create(
         vector_store_id: str,
         *,
         file_id: str,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -62,7 +72,7 @@ def create(
               files.
 
           chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
-              strategy.
+              strategy. Only applicable if `file_ids` is non-empty.
 
           extra_headers: Send extra headers
 
@@ -154,8 +164,8 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`.
 
@@ -245,7 +255,7 @@ def create_and_poll(
         *,
         vector_store_id: str,
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Attach a file to the given vector store and wait for it to be processed."""
         self.create(vector_store_id=vector_store_id, file_id=file_id, chunking_strategy=chunking_strategy)
@@ -302,7 +312,7 @@ def upload(
         *,
         vector_store_id: str,
         file: FileTypes,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Upload a file to the `files` API and then attach it to the given vector store.
 
@@ -318,7 +328,7 @@ def upload_and_poll(
         vector_store_id: str,
         file: FileTypes,
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Add a file to a vector store and poll until processing is complete."""
         file_obj = self._client.files.create(file=file, purpose="assistants")
@@ -333,10 +343,21 @@ def upload_and_poll(
 class AsyncFiles(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncFilesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncFilesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncFilesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncFilesWithStreamingResponse(self)
 
     async def create(
@@ -344,7 +365,7 @@ async def create(
         vector_store_id: str,
         *,
         file_id: str,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -363,7 +384,7 @@ async def create(
               files.
 
           chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
-              strategy.
+              strategy. Only applicable if `file_ids` is non-empty.
 
           extra_headers: Send extra headers
 
@@ -455,8 +476,8 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`.
 
@@ -546,7 +567,7 @@ async def create_and_poll(
         *,
         vector_store_id: str,
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Attach a file to the given vector store and wait for it to be processed."""
         await self.create(vector_store_id=vector_store_id, file_id=file_id, chunking_strategy=chunking_strategy)
@@ -603,7 +624,7 @@ async def upload(
         *,
         vector_store_id: str,
         file: FileTypes,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Upload a file to the `files` API and then attach it to the given vector store.
 
@@ -621,7 +642,7 @@ async def upload_and_poll(
         vector_store_id: str,
         file: FileTypes,
         poll_interval_ms: int | NotGiven = NOT_GIVEN,
-        chunking_strategy: file_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
     ) -> VectorStoreFile:
         """Add a file to a vector store and poll until processing is complete."""
         file_obj = await self._client.files.create(file=file, purpose="assistants")
diff --git a/portkey_ai/_vendor/openai/resources/beta/vector_stores/vector_stores.py b/portkey_ai/_vendor/openai/resources/beta/vector_stores/vector_stores.py
index cbd56a06..61a2eadc 100644
--- a/portkey_ai/_vendor/openai/resources/beta/vector_stores/vector_stores.py
+++ b/portkey_ai/_vendor/openai/resources/beta/vector_stores/vector_stores.py
@@ -33,13 +33,16 @@
     AsyncFileBatchesWithStreamingResponse,
 )
 from ....pagination import SyncCursorPage, AsyncCursorPage
-from ....types.beta import vector_store_list_params, vector_store_create_params, vector_store_update_params
-from ...._base_client import (
-    AsyncPaginator,
-    make_request_options,
+from ....types.beta import (
+    FileChunkingStrategyParam,
+    vector_store_list_params,
+    vector_store_create_params,
+    vector_store_update_params,
 )
+from ...._base_client import AsyncPaginator, make_request_options
 from ....types.beta.vector_store import VectorStore
 from ....types.beta.vector_store_deleted import VectorStoreDeleted
+from ....types.beta.file_chunking_strategy_param import FileChunkingStrategyParam
 
 __all__ = ["VectorStores", "AsyncVectorStores"]
 
@@ -55,16 +58,27 @@ def file_batches(self) -> FileBatches:
 
     @cached_property
     def with_raw_response(self) -> VectorStoresWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return VectorStoresWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> VectorStoresWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return VectorStoresWithStreamingResponse(self)
 
     def create(
         self,
         *,
-        chunking_strategy: vector_store_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         expires_after: vector_store_create_params.ExpiresAfter | NotGiven = NOT_GIVEN,
         file_ids: List[str] | NotGiven = NOT_GIVEN,
         metadata: Optional[object] | NotGiven = NOT_GIVEN,
@@ -91,7 +105,7 @@ def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           name: The name of the vector store.
@@ -179,7 +193,7 @@ def update(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           name: The name of the vector store.
@@ -237,8 +251,8 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           limit: A limit on the number of objects to be returned. Limit can range between 1 and
               100, and the default is 20.
@@ -322,16 +336,27 @@ def file_batches(self) -> AsyncFileBatches:
 
     @cached_property
     def with_raw_response(self) -> AsyncVectorStoresWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncVectorStoresWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncVectorStoresWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncVectorStoresWithStreamingResponse(self)
 
     async def create(
         self,
         *,
-        chunking_strategy: vector_store_create_params.ChunkingStrategy | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
         expires_after: vector_store_create_params.ExpiresAfter | NotGiven = NOT_GIVEN,
         file_ids: List[str] | NotGiven = NOT_GIVEN,
         metadata: Optional[object] | NotGiven = NOT_GIVEN,
@@ -358,7 +383,7 @@ async def create(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           name: The name of the vector store.
@@ -446,7 +471,7 @@ async def update(
 
           metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
               for storing additional information about the object in a structured format. Keys
-              can be a maximum of 64 characters long and values can be a maxium of 512
+              can be a maximum of 64 characters long and values can be a maximum of 512
               characters long.
 
           name: The name of the vector store.
@@ -504,8 +529,8 @@ def list(
 
           before: A cursor for use in pagination. `before` is an object ID that defines your place
               in the list. For instance, if you make a list request and receive 100 objects,
-              ending with obj_foo, your subsequent call can include before=obj_foo in order to
-              fetch the previous page of the list.
+              starting with obj_foo, your subsequent call can include before=obj_foo in order
+              to fetch the previous page of the list.
 
           limit: A limit on the number of objects to be returned. Limit can range between 1 and
               100, and the default is 20.
diff --git a/portkey_ai/_vendor/openai/resources/chat/chat.py b/portkey_ai/_vendor/openai/resources/chat/chat.py
index d14d0555..dc23a15a 100644
--- a/portkey_ai/_vendor/openai/resources/chat/chat.py
+++ b/portkey_ai/_vendor/openai/resources/chat/chat.py
@@ -23,10 +23,21 @@ def completions(self) -> Completions:
 
     @cached_property
     def with_raw_response(self) -> ChatWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return ChatWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> ChatWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return ChatWithStreamingResponse(self)
 
 
@@ -37,10 +48,21 @@ def completions(self) -> AsyncCompletions:
 
     @cached_property
     def with_raw_response(self) -> AsyncChatWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncChatWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncChatWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncChatWithStreamingResponse(self)
 
 
diff --git a/portkey_ai/_vendor/openai/resources/chat/completions.py b/portkey_ai/_vendor/openai/resources/chat/completions.py
index 3dcd3774..728c7443 100644
--- a/portkey_ai/_vendor/openai/resources/chat/completions.py
+++ b/portkey_ai/_vendor/openai/resources/chat/completions.py
@@ -2,10 +2,12 @@
 
 from __future__ import annotations
 
-from typing import Dict, List, Union, Iterable, Optional, overload
-from typing_extensions import Literal
+import inspect
+from typing import Dict, List, Union, Iterable, Optional
+from typing_extensions import Literal, overload
 
 import httpx
+import pydantic
 
 from ... import _legacy_response
 from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven
@@ -18,14 +20,22 @@
 from ..._resource import SyncAPIResource, AsyncAPIResource
 from ..._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
 from ..._streaming import Stream, AsyncStream
-from ...types.chat import completion_create_params
+from ...types.chat import (
+    ChatCompletionAudioParam,
+    ChatCompletionReasoningEffort,
+    completion_create_params,
+)
 from ..._base_client import make_request_options
 from ...types.chat_model import ChatModel
 from ...types.chat.chat_completion import ChatCompletion
 from ...types.chat.chat_completion_chunk import ChatCompletionChunk
+from ...types.chat.chat_completion_modality import ChatCompletionModality
 from ...types.chat.chat_completion_tool_param import ChatCompletionToolParam
+from ...types.chat.chat_completion_audio_param import ChatCompletionAudioParam
 from ...types.chat.chat_completion_message_param import ChatCompletionMessageParam
+from ...types.chat.chat_completion_reasoning_effort import ChatCompletionReasoningEffort
 from ...types.chat.chat_completion_stream_options_param import ChatCompletionStreamOptionsParam
+from ...types.chat.chat_completion_prediction_content_param import ChatCompletionPredictionContentParam
 from ...types.chat.chat_completion_tool_choice_option_param import ChatCompletionToolChoiceOptionParam
 
 __all__ = ["Completions", "AsyncCompletions"]
@@ -34,10 +44,21 @@
 class Completions(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> CompletionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return CompletionsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> CompletionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return CompletionsWithStreamingResponse(self)
 
     @overload
@@ -46,19 +67,26 @@ def create(
         *,
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         response_format: completion_create_params.ResponseFormat | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream: Optional[Literal[False]] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
@@ -74,31 +102,51 @@ def create(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> ChatCompletion:
-        """
-        Creates a model response for the given chat conversation.
+        """Creates a model response for the given chat conversation.
+
+        Learn more in the
+        [text generation](https://platform.openai.com/docs/guides/text-generation),
+        [vision](https://platform.openai.com/docs/guides/vision), and
+        [audio](https://platform.openai.com/docs/guides/audio) guides.
+
+        Parameter support can differ depending on the model used to generate the
+        response, particularly for newer reasoning models. Parameters that are only
+        supported for reasoning models are noted below. For the current state of
+        unsupported parameters in reasoning models,
+        [refer to the reasoning guide](https://platform.openai.com/docs/guides/reasoning).
 
         Args:
-          messages: A list of messages comprising the conversation so far.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_format_inputs_to_chatgpt_models).
+          messages: A list of messages comprising the conversation so far. Depending on the
+              [model](https://platform.openai.com/docs/models) you use, different message
+              types (modalities) are supported, like
+              [text](https://platform.openai.com/docs/guides/text-generation),
+              [images](https://platform.openai.com/docs/guides/vision), and
+              [audio](https://platform.openai.com/docs/guides/audio).
 
           model: ID of the model to use. See the
-              [model endpoint compatibility](https://platform.openai.com/docs/models/model-endpoint-compatibility)
+              [model endpoint compatibility](https://platform.openai.com/docs/models#model-endpoint-compatibility)
               table for details on which models work with the Chat API.
 
+          audio: Parameters for audio output. Required when audio output is requested with
+              `modalities: ["audio"]`.
+              [Learn more](https://platform.openai.com/docs/guides/audio).
+
           frequency_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on their
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
-
           function_call: Deprecated in favor of `tool_choice`.
 
-              Controls which (if any) function is called by the model. `none` means the model
-              will not call a function and instead generates a message. `auto` means the model
-              can pick between generating a message or calling a function. Specifying a
-              particular function via `{"name": "my_function"}` forces the model to call that
+              Controls which (if any) function is called by the model.
+
+              `none` means the model will not call a function and instead generates a message.
+
+              `auto` means the model can pick between generating a message or calling a
               function.
 
+              Specifying a particular function via `{"name": "my_function"}` forces the model
+              to call that function.
+
               `none` is the default when no functions are present. `auto` is the default if
               functions are present.
 
@@ -119,35 +167,62 @@ def create(
               returns the log probabilities of each output token returned in the `content` of
               `message`.
 
+          max_completion_tokens: An upper bound for the number of tokens that can be generated for a completion,
+              including visible output tokens and
+              [reasoning tokens](https://platform.openai.com/docs/guides/reasoning).
+
           max_tokens: The maximum number of [tokens](/tokenizer) that can be generated in the chat
-              completion.
+              completion. This value can be used to control
+              [costs](https://openai.com/api/pricing/) for text generated via API.
+
+              This value is now deprecated in favor of `max_completion_tokens`, and is not
+              compatible with
+              [o1 series models](https://platform.openai.com/docs/guides/reasoning).
 
-              The total length of input tokens and generated tokens is limited by the model's
-              context length.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken)
-              for counting tokens.
+          metadata: Developer-defined tags and values used for filtering completions in the
+              [dashboard](https://platform.openai.com/chat-completions).
+
+          modalities: Output types that you would like the model to generate for this request. Most
+              models are capable of generating text, which is the default:
+
+              `["text"]`
+
+              The `gpt-4o-audio-preview` model can also be used to
+              [generate audio](https://platform.openai.com/docs/guides/audio). To request that
+              this model generate both text and audio responses, you can use:
+
+              `["text", "audio"]`
 
           n: How many chat completion choices to generate for each input message. Note that
               you will be charged based on the number of generated tokens across all of the
               choices. Keep `n` as `1` to minimize costs.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
+          prediction: Static predicted output content, such as the content of a text file that is
+              being regenerated.
+
           presence_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+          reasoning_effort: **o1 models only**
+
+              Constrains effort on reasoning for
+              [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+              supported values are `low`, `medium`, and `high`. Reducing reasoning effort can
+              result in faster responses and fewer tokens used on reasoning in a response.
 
-          response_format: An object specifying the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4o mini](https://platform.openai.com/docs/models/gpt-4o-mini),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo) and
-              all GPT-3.5 Turbo models newer than `gpt-3.5-turbo-1106`.
+          response_format: An object specifying the format that the model must output.
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
+              [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
+
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -167,8 +242,11 @@ def create(
           service_tier: Specifies the latency tier to use for processing the request. This parameter is
               relevant for customers subscribed to the scale tier service:
 
-              - If set to 'auto', the system will utilize scale tier credits until they are
-                exhausted.
+              - If set to 'auto', and the Project is Scale tier enabled, the system will
+                utilize scale tier credits until they are exhausted.
+              - If set to 'auto', and the Project is not Scale tier enabled, the request will
+                be processed using the default service tier with a lower uptime SLA and no
+                latency guarentee.
               - If set to 'default', the request will be processed using the default service
                 tier with a lower uptime SLA and no latency guarentee.
               - When not set, the default behavior is 'auto'.
@@ -178,6 +256,10 @@ def create(
 
           stop: Up to 4 sequences where the API will stop generating further tokens.
 
+          store: Whether or not to store the output of this chat completion request for use in
+              our [model distillation](https://platform.openai.com/docs/guides/distillation)
+              or [evals](https://platform.openai.com/docs/guides/evals) products.
+
           stream: If set, partial message deltas will be sent, like in ChatGPT. Tokens will be
               sent as data-only
               [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format)
@@ -189,9 +271,8 @@ def create(
 
           temperature: What sampling temperature to use, between 0 and 2. Higher values like 0.8 will
               make the output more random, while lower values like 0.2 will make it more
-              focused and deterministic.
-
-              We generally recommend altering this or `top_p` but not both.
+              focused and deterministic. We generally recommend altering this or `top_p` but
+              not both.
 
           tool_choice: Controls which (if any) tool is called by the model. `none` means the model will
               not call any tool and instead generates a message. `auto` means the model can
@@ -219,7 +300,7 @@ def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -238,19 +319,26 @@ def create(
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
         stream: Literal[True],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         response_format: completion_create_params.ResponseFormat | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
@@ -265,15 +353,29 @@ def create(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Stream[ChatCompletionChunk]:
-        """
-        Creates a model response for the given chat conversation.
+        """Creates a model response for the given chat conversation.
+
+        Learn more in the
+        [text generation](https://platform.openai.com/docs/guides/text-generation),
+        [vision](https://platform.openai.com/docs/guides/vision), and
+        [audio](https://platform.openai.com/docs/guides/audio) guides.
+
+        Parameter support can differ depending on the model used to generate the
+        response, particularly for newer reasoning models. Parameters that are only
+        supported for reasoning models are noted below. For the current state of
+        unsupported parameters in reasoning models,
+        [refer to the reasoning guide](https://platform.openai.com/docs/guides/reasoning).
 
         Args:
-          messages: A list of messages comprising the conversation so far.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_format_inputs_to_chatgpt_models).
+          messages: A list of messages comprising the conversation so far. Depending on the
+              [model](https://platform.openai.com/docs/models) you use, different message
+              types (modalities) are supported, like
+              [text](https://platform.openai.com/docs/guides/text-generation),
+              [images](https://platform.openai.com/docs/guides/vision), and
+              [audio](https://platform.openai.com/docs/guides/audio).
 
           model: ID of the model to use. See the
-              [model endpoint compatibility](https://platform.openai.com/docs/models/model-endpoint-compatibility)
+              [model endpoint compatibility](https://platform.openai.com/docs/models#model-endpoint-compatibility)
               table for details on which models work with the Chat API.
 
           stream: If set, partial message deltas will be sent, like in ChatGPT. Tokens will be
@@ -283,20 +385,26 @@ def create(
               message.
               [Example Python code](https://cookbook.openai.com/examples/how_to_stream_completions).
 
+          audio: Parameters for audio output. Required when audio output is requested with
+              `modalities: ["audio"]`.
+              [Learn more](https://platform.openai.com/docs/guides/audio).
+
           frequency_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on their
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
-
           function_call: Deprecated in favor of `tool_choice`.
 
-              Controls which (if any) function is called by the model. `none` means the model
-              will not call a function and instead generates a message. `auto` means the model
-              can pick between generating a message or calling a function. Specifying a
-              particular function via `{"name": "my_function"}` forces the model to call that
+              Controls which (if any) function is called by the model.
+
+              `none` means the model will not call a function and instead generates a message.
+
+              `auto` means the model can pick between generating a message or calling a
               function.
 
+              Specifying a particular function via `{"name": "my_function"}` forces the model
+              to call that function.
+
               `none` is the default when no functions are present. `auto` is the default if
               functions are present.
 
@@ -317,35 +425,62 @@ def create(
               returns the log probabilities of each output token returned in the `content` of
               `message`.
 
+          max_completion_tokens: An upper bound for the number of tokens that can be generated for a completion,
+              including visible output tokens and
+              [reasoning tokens](https://platform.openai.com/docs/guides/reasoning).
+
           max_tokens: The maximum number of [tokens](/tokenizer) that can be generated in the chat
-              completion.
+              completion. This value can be used to control
+              [costs](https://openai.com/api/pricing/) for text generated via API.
+
+              This value is now deprecated in favor of `max_completion_tokens`, and is not
+              compatible with
+              [o1 series models](https://platform.openai.com/docs/guides/reasoning).
+
+          metadata: Developer-defined tags and values used for filtering completions in the
+              [dashboard](https://platform.openai.com/chat-completions).
 
-              The total length of input tokens and generated tokens is limited by the model's
-              context length.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken)
-              for counting tokens.
+          modalities: Output types that you would like the model to generate for this request. Most
+              models are capable of generating text, which is the default:
+
+              `["text"]`
+
+              The `gpt-4o-audio-preview` model can also be used to
+              [generate audio](https://platform.openai.com/docs/guides/audio). To request that
+              this model generate both text and audio responses, you can use:
+
+              `["text", "audio"]`
 
           n: How many chat completion choices to generate for each input message. Note that
               you will be charged based on the number of generated tokens across all of the
               choices. Keep `n` as `1` to minimize costs.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
+          prediction: Static predicted output content, such as the content of a text file that is
+              being regenerated.
+
           presence_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+          reasoning_effort: **o1 models only**
 
-          response_format: An object specifying the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4o mini](https://platform.openai.com/docs/models/gpt-4o-mini),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo) and
-              all GPT-3.5 Turbo models newer than `gpt-3.5-turbo-1106`.
+              Constrains effort on reasoning for
+              [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+              supported values are `low`, `medium`, and `high`. Reducing reasoning effort can
+              result in faster responses and fewer tokens used on reasoning in a response.
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+          response_format: An object specifying the format that the model must output.
+
+              Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
+              [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
+
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -365,8 +500,11 @@ def create(
           service_tier: Specifies the latency tier to use for processing the request. This parameter is
               relevant for customers subscribed to the scale tier service:
 
-              - If set to 'auto', the system will utilize scale tier credits until they are
-                exhausted.
+              - If set to 'auto', and the Project is Scale tier enabled, the system will
+                utilize scale tier credits until they are exhausted.
+              - If set to 'auto', and the Project is not Scale tier enabled, the request will
+                be processed using the default service tier with a lower uptime SLA and no
+                latency guarentee.
               - If set to 'default', the request will be processed using the default service
                 tier with a lower uptime SLA and no latency guarentee.
               - When not set, the default behavior is 'auto'.
@@ -376,13 +514,16 @@ def create(
 
           stop: Up to 4 sequences where the API will stop generating further tokens.
 
+          store: Whether or not to store the output of this chat completion request for use in
+              our [model distillation](https://platform.openai.com/docs/guides/distillation)
+              or [evals](https://platform.openai.com/docs/guides/evals) products.
+
           stream_options: Options for streaming response. Only set this when you set `stream: true`.
 
           temperature: What sampling temperature to use, between 0 and 2. Higher values like 0.8 will
               make the output more random, while lower values like 0.2 will make it more
-              focused and deterministic.
-
-              We generally recommend altering this or `top_p` but not both.
+              focused and deterministic. We generally recommend altering this or `top_p` but
+              not both.
 
           tool_choice: Controls which (if any) tool is called by the model. `none` means the model will
               not call any tool and instead generates a message. `auto` means the model can
@@ -410,7 +551,7 @@ def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -429,19 +570,26 @@ def create(
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
         stream: bool,
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         response_format: completion_create_params.ResponseFormat | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
@@ -456,15 +604,29 @@ def create(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> ChatCompletion | Stream[ChatCompletionChunk]:
-        """
-        Creates a model response for the given chat conversation.
+        """Creates a model response for the given chat conversation.
+
+        Learn more in the
+        [text generation](https://platform.openai.com/docs/guides/text-generation),
+        [vision](https://platform.openai.com/docs/guides/vision), and
+        [audio](https://platform.openai.com/docs/guides/audio) guides.
+
+        Parameter support can differ depending on the model used to generate the
+        response, particularly for newer reasoning models. Parameters that are only
+        supported for reasoning models are noted below. For the current state of
+        unsupported parameters in reasoning models,
+        [refer to the reasoning guide](https://platform.openai.com/docs/guides/reasoning).
 
         Args:
-          messages: A list of messages comprising the conversation so far.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_format_inputs_to_chatgpt_models).
+          messages: A list of messages comprising the conversation so far. Depending on the
+              [model](https://platform.openai.com/docs/models) you use, different message
+              types (modalities) are supported, like
+              [text](https://platform.openai.com/docs/guides/text-generation),
+              [images](https://platform.openai.com/docs/guides/vision), and
+              [audio](https://platform.openai.com/docs/guides/audio).
 
           model: ID of the model to use. See the
-              [model endpoint compatibility](https://platform.openai.com/docs/models/model-endpoint-compatibility)
+              [model endpoint compatibility](https://platform.openai.com/docs/models#model-endpoint-compatibility)
               table for details on which models work with the Chat API.
 
           stream: If set, partial message deltas will be sent, like in ChatGPT. Tokens will be
@@ -474,20 +636,26 @@ def create(
               message.
               [Example Python code](https://cookbook.openai.com/examples/how_to_stream_completions).
 
+          audio: Parameters for audio output. Required when audio output is requested with
+              `modalities: ["audio"]`.
+              [Learn more](https://platform.openai.com/docs/guides/audio).
+
           frequency_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on their
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
-
           function_call: Deprecated in favor of `tool_choice`.
 
-              Controls which (if any) function is called by the model. `none` means the model
-              will not call a function and instead generates a message. `auto` means the model
-              can pick between generating a message or calling a function. Specifying a
-              particular function via `{"name": "my_function"}` forces the model to call that
+              Controls which (if any) function is called by the model.
+
+              `none` means the model will not call a function and instead generates a message.
+
+              `auto` means the model can pick between generating a message or calling a
               function.
 
+              Specifying a particular function via `{"name": "my_function"}` forces the model
+              to call that function.
+
               `none` is the default when no functions are present. `auto` is the default if
               functions are present.
 
@@ -508,35 +676,62 @@ def create(
               returns the log probabilities of each output token returned in the `content` of
               `message`.
 
+          max_completion_tokens: An upper bound for the number of tokens that can be generated for a completion,
+              including visible output tokens and
+              [reasoning tokens](https://platform.openai.com/docs/guides/reasoning).
+
           max_tokens: The maximum number of [tokens](/tokenizer) that can be generated in the chat
-              completion.
+              completion. This value can be used to control
+              [costs](https://openai.com/api/pricing/) for text generated via API.
+
+              This value is now deprecated in favor of `max_completion_tokens`, and is not
+              compatible with
+              [o1 series models](https://platform.openai.com/docs/guides/reasoning).
+
+          metadata: Developer-defined tags and values used for filtering completions in the
+              [dashboard](https://platform.openai.com/chat-completions).
+
+          modalities: Output types that you would like the model to generate for this request. Most
+              models are capable of generating text, which is the default:
 
-              The total length of input tokens and generated tokens is limited by the model's
-              context length.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken)
-              for counting tokens.
+              `["text"]`
+
+              The `gpt-4o-audio-preview` model can also be used to
+              [generate audio](https://platform.openai.com/docs/guides/audio). To request that
+              this model generate both text and audio responses, you can use:
+
+              `["text", "audio"]`
 
           n: How many chat completion choices to generate for each input message. Note that
               you will be charged based on the number of generated tokens across all of the
               choices. Keep `n` as `1` to minimize costs.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
+          prediction: Static predicted output content, such as the content of a text file that is
+              being regenerated.
+
           presence_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+          reasoning_effort: **o1 models only**
+
+              Constrains effort on reasoning for
+              [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+              supported values are `low`, `medium`, and `high`. Reducing reasoning effort can
+              result in faster responses and fewer tokens used on reasoning in a response.
 
-          response_format: An object specifying the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4o mini](https://platform.openai.com/docs/models/gpt-4o-mini),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo) and
-              all GPT-3.5 Turbo models newer than `gpt-3.5-turbo-1106`.
+          response_format: An object specifying the format that the model must output.
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
+              [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
+
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -556,8 +751,11 @@ def create(
           service_tier: Specifies the latency tier to use for processing the request. This parameter is
               relevant for customers subscribed to the scale tier service:
 
-              - If set to 'auto', the system will utilize scale tier credits until they are
-                exhausted.
+              - If set to 'auto', and the Project is Scale tier enabled, the system will
+                utilize scale tier credits until they are exhausted.
+              - If set to 'auto', and the Project is not Scale tier enabled, the request will
+                be processed using the default service tier with a lower uptime SLA and no
+                latency guarentee.
               - If set to 'default', the request will be processed using the default service
                 tier with a lower uptime SLA and no latency guarentee.
               - When not set, the default behavior is 'auto'.
@@ -567,13 +765,16 @@ def create(
 
           stop: Up to 4 sequences where the API will stop generating further tokens.
 
+          store: Whether or not to store the output of this chat completion request for use in
+              our [model distillation](https://platform.openai.com/docs/guides/distillation)
+              or [evals](https://platform.openai.com/docs/guides/evals) products.
+
           stream_options: Options for streaming response. Only set this when you set `stream: true`.
 
           temperature: What sampling temperature to use, between 0 and 2. Higher values like 0.8 will
               make the output more random, while lower values like 0.2 will make it more
-              focused and deterministic.
-
-              We generally recommend altering this or `top_p` but not both.
+              focused and deterministic. We generally recommend altering this or `top_p` but
+              not both.
 
           tool_choice: Controls which (if any) tool is called by the model. `none` means the model will
               not call any tool and instead generates a message. `auto` means the model can
@@ -601,7 +802,7 @@ def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -619,19 +820,26 @@ def create(
         *,
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         response_format: completion_create_params.ResponseFormat | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream: Optional[Literal[False]] | Literal[True] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
@@ -647,25 +855,33 @@ def create(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> ChatCompletion | Stream[ChatCompletionChunk]:
+        validate_response_format(response_format)
         return self._post(
             "/chat/completions",
             body=maybe_transform(
                 {
                     "messages": messages,
                     "model": model,
+                    "audio": audio,
                     "frequency_penalty": frequency_penalty,
                     "function_call": function_call,
                     "functions": functions,
                     "logit_bias": logit_bias,
                     "logprobs": logprobs,
+                    "max_completion_tokens": max_completion_tokens,
                     "max_tokens": max_tokens,
+                    "metadata": metadata,
+                    "modalities": modalities,
                     "n": n,
                     "parallel_tool_calls": parallel_tool_calls,
+                    "prediction": prediction,
                     "presence_penalty": presence_penalty,
+                    "reasoning_effort": reasoning_effort,
                     "response_format": response_format,
                     "seed": seed,
                     "service_tier": service_tier,
                     "stop": stop,
+                    "store": store,
                     "stream": stream,
                     "stream_options": stream_options,
                     "temperature": temperature,
@@ -689,10 +905,21 @@ def create(
 class AsyncCompletions(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncCompletionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncCompletionsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncCompletionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncCompletionsWithStreamingResponse(self)
 
     @overload
@@ -701,19 +928,26 @@ async def create(
         *,
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         response_format: completion_create_params.ResponseFormat | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream: Optional[Literal[False]] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
@@ -729,31 +963,51 @@ async def create(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> ChatCompletion:
-        """
-        Creates a model response for the given chat conversation.
+        """Creates a model response for the given chat conversation.
+
+        Learn more in the
+        [text generation](https://platform.openai.com/docs/guides/text-generation),
+        [vision](https://platform.openai.com/docs/guides/vision), and
+        [audio](https://platform.openai.com/docs/guides/audio) guides.
+
+        Parameter support can differ depending on the model used to generate the
+        response, particularly for newer reasoning models. Parameters that are only
+        supported for reasoning models are noted below. For the current state of
+        unsupported parameters in reasoning models,
+        [refer to the reasoning guide](https://platform.openai.com/docs/guides/reasoning).
 
         Args:
-          messages: A list of messages comprising the conversation so far.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_format_inputs_to_chatgpt_models).
+          messages: A list of messages comprising the conversation so far. Depending on the
+              [model](https://platform.openai.com/docs/models) you use, different message
+              types (modalities) are supported, like
+              [text](https://platform.openai.com/docs/guides/text-generation),
+              [images](https://platform.openai.com/docs/guides/vision), and
+              [audio](https://platform.openai.com/docs/guides/audio).
 
           model: ID of the model to use. See the
-              [model endpoint compatibility](https://platform.openai.com/docs/models/model-endpoint-compatibility)
+              [model endpoint compatibility](https://platform.openai.com/docs/models#model-endpoint-compatibility)
               table for details on which models work with the Chat API.
 
+          audio: Parameters for audio output. Required when audio output is requested with
+              `modalities: ["audio"]`.
+              [Learn more](https://platform.openai.com/docs/guides/audio).
+
           frequency_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on their
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
-
           function_call: Deprecated in favor of `tool_choice`.
 
-              Controls which (if any) function is called by the model. `none` means the model
-              will not call a function and instead generates a message. `auto` means the model
-              can pick between generating a message or calling a function. Specifying a
-              particular function via `{"name": "my_function"}` forces the model to call that
+              Controls which (if any) function is called by the model.
+
+              `none` means the model will not call a function and instead generates a message.
+
+              `auto` means the model can pick between generating a message or calling a
               function.
 
+              Specifying a particular function via `{"name": "my_function"}` forces the model
+              to call that function.
+
               `none` is the default when no functions are present. `auto` is the default if
               functions are present.
 
@@ -774,35 +1028,62 @@ async def create(
               returns the log probabilities of each output token returned in the `content` of
               `message`.
 
+          max_completion_tokens: An upper bound for the number of tokens that can be generated for a completion,
+              including visible output tokens and
+              [reasoning tokens](https://platform.openai.com/docs/guides/reasoning).
+
           max_tokens: The maximum number of [tokens](/tokenizer) that can be generated in the chat
-              completion.
+              completion. This value can be used to control
+              [costs](https://openai.com/api/pricing/) for text generated via API.
+
+              This value is now deprecated in favor of `max_completion_tokens`, and is not
+              compatible with
+              [o1 series models](https://platform.openai.com/docs/guides/reasoning).
+
+          metadata: Developer-defined tags and values used for filtering completions in the
+              [dashboard](https://platform.openai.com/chat-completions).
+
+          modalities: Output types that you would like the model to generate for this request. Most
+              models are capable of generating text, which is the default:
 
-              The total length of input tokens and generated tokens is limited by the model's
-              context length.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken)
-              for counting tokens.
+              `["text"]`
+
+              The `gpt-4o-audio-preview` model can also be used to
+              [generate audio](https://platform.openai.com/docs/guides/audio). To request that
+              this model generate both text and audio responses, you can use:
+
+              `["text", "audio"]`
 
           n: How many chat completion choices to generate for each input message. Note that
               you will be charged based on the number of generated tokens across all of the
               choices. Keep `n` as `1` to minimize costs.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
+          prediction: Static predicted output content, such as the content of a text file that is
+              being regenerated.
+
           presence_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+          reasoning_effort: **o1 models only**
+
+              Constrains effort on reasoning for
+              [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+              supported values are `low`, `medium`, and `high`. Reducing reasoning effort can
+              result in faster responses and fewer tokens used on reasoning in a response.
 
-          response_format: An object specifying the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4o mini](https://platform.openai.com/docs/models/gpt-4o-mini),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo) and
-              all GPT-3.5 Turbo models newer than `gpt-3.5-turbo-1106`.
+          response_format: An object specifying the format that the model must output.
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
+              [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
+
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -822,8 +1103,11 @@ async def create(
           service_tier: Specifies the latency tier to use for processing the request. This parameter is
               relevant for customers subscribed to the scale tier service:
 
-              - If set to 'auto', the system will utilize scale tier credits until they are
-                exhausted.
+              - If set to 'auto', and the Project is Scale tier enabled, the system will
+                utilize scale tier credits until they are exhausted.
+              - If set to 'auto', and the Project is not Scale tier enabled, the request will
+                be processed using the default service tier with a lower uptime SLA and no
+                latency guarentee.
               - If set to 'default', the request will be processed using the default service
                 tier with a lower uptime SLA and no latency guarentee.
               - When not set, the default behavior is 'auto'.
@@ -833,6 +1117,10 @@ async def create(
 
           stop: Up to 4 sequences where the API will stop generating further tokens.
 
+          store: Whether or not to store the output of this chat completion request for use in
+              our [model distillation](https://platform.openai.com/docs/guides/distillation)
+              or [evals](https://platform.openai.com/docs/guides/evals) products.
+
           stream: If set, partial message deltas will be sent, like in ChatGPT. Tokens will be
               sent as data-only
               [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format)
@@ -844,9 +1132,8 @@ async def create(
 
           temperature: What sampling temperature to use, between 0 and 2. Higher values like 0.8 will
               make the output more random, while lower values like 0.2 will make it more
-              focused and deterministic.
-
-              We generally recommend altering this or `top_p` but not both.
+              focused and deterministic. We generally recommend altering this or `top_p` but
+              not both.
 
           tool_choice: Controls which (if any) tool is called by the model. `none` means the model will
               not call any tool and instead generates a message. `auto` means the model can
@@ -874,7 +1161,7 @@ async def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -893,19 +1180,26 @@ async def create(
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
         stream: Literal[True],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         response_format: completion_create_params.ResponseFormat | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
@@ -920,15 +1214,29 @@ async def create(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> AsyncStream[ChatCompletionChunk]:
-        """
-        Creates a model response for the given chat conversation.
+        """Creates a model response for the given chat conversation.
+
+        Learn more in the
+        [text generation](https://platform.openai.com/docs/guides/text-generation),
+        [vision](https://platform.openai.com/docs/guides/vision), and
+        [audio](https://platform.openai.com/docs/guides/audio) guides.
+
+        Parameter support can differ depending on the model used to generate the
+        response, particularly for newer reasoning models. Parameters that are only
+        supported for reasoning models are noted below. For the current state of
+        unsupported parameters in reasoning models,
+        [refer to the reasoning guide](https://platform.openai.com/docs/guides/reasoning).
 
         Args:
-          messages: A list of messages comprising the conversation so far.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_format_inputs_to_chatgpt_models).
+          messages: A list of messages comprising the conversation so far. Depending on the
+              [model](https://platform.openai.com/docs/models) you use, different message
+              types (modalities) are supported, like
+              [text](https://platform.openai.com/docs/guides/text-generation),
+              [images](https://platform.openai.com/docs/guides/vision), and
+              [audio](https://platform.openai.com/docs/guides/audio).
 
           model: ID of the model to use. See the
-              [model endpoint compatibility](https://platform.openai.com/docs/models/model-endpoint-compatibility)
+              [model endpoint compatibility](https://platform.openai.com/docs/models#model-endpoint-compatibility)
               table for details on which models work with the Chat API.
 
           stream: If set, partial message deltas will be sent, like in ChatGPT. Tokens will be
@@ -938,20 +1246,26 @@ async def create(
               message.
               [Example Python code](https://cookbook.openai.com/examples/how_to_stream_completions).
 
+          audio: Parameters for audio output. Required when audio output is requested with
+              `modalities: ["audio"]`.
+              [Learn more](https://platform.openai.com/docs/guides/audio).
+
           frequency_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on their
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
-
           function_call: Deprecated in favor of `tool_choice`.
 
-              Controls which (if any) function is called by the model. `none` means the model
-              will not call a function and instead generates a message. `auto` means the model
-              can pick between generating a message or calling a function. Specifying a
-              particular function via `{"name": "my_function"}` forces the model to call that
+              Controls which (if any) function is called by the model.
+
+              `none` means the model will not call a function and instead generates a message.
+
+              `auto` means the model can pick between generating a message or calling a
               function.
 
+              Specifying a particular function via `{"name": "my_function"}` forces the model
+              to call that function.
+
               `none` is the default when no functions are present. `auto` is the default if
               functions are present.
 
@@ -972,35 +1286,62 @@ async def create(
               returns the log probabilities of each output token returned in the `content` of
               `message`.
 
+          max_completion_tokens: An upper bound for the number of tokens that can be generated for a completion,
+              including visible output tokens and
+              [reasoning tokens](https://platform.openai.com/docs/guides/reasoning).
+
           max_tokens: The maximum number of [tokens](/tokenizer) that can be generated in the chat
-              completion.
+              completion. This value can be used to control
+              [costs](https://openai.com/api/pricing/) for text generated via API.
 
-              The total length of input tokens and generated tokens is limited by the model's
-              context length.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken)
-              for counting tokens.
+              This value is now deprecated in favor of `max_completion_tokens`, and is not
+              compatible with
+              [o1 series models](https://platform.openai.com/docs/guides/reasoning).
+
+          metadata: Developer-defined tags and values used for filtering completions in the
+              [dashboard](https://platform.openai.com/chat-completions).
+
+          modalities: Output types that you would like the model to generate for this request. Most
+              models are capable of generating text, which is the default:
+
+              `["text"]`
+
+              The `gpt-4o-audio-preview` model can also be used to
+              [generate audio](https://platform.openai.com/docs/guides/audio). To request that
+              this model generate both text and audio responses, you can use:
+
+              `["text", "audio"]`
 
           n: How many chat completion choices to generate for each input message. Note that
               you will be charged based on the number of generated tokens across all of the
               choices. Keep `n` as `1` to minimize costs.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
+          prediction: Static predicted output content, such as the content of a text file that is
+              being regenerated.
+
           presence_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+          reasoning_effort: **o1 models only**
+
+              Constrains effort on reasoning for
+              [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+              supported values are `low`, `medium`, and `high`. Reducing reasoning effort can
+              result in faster responses and fewer tokens used on reasoning in a response.
+
+          response_format: An object specifying the format that the model must output.
 
-          response_format: An object specifying the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4o mini](https://platform.openai.com/docs/models/gpt-4o-mini),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo) and
-              all GPT-3.5 Turbo models newer than `gpt-3.5-turbo-1106`.
+              Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
+              [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -1020,8 +1361,11 @@ async def create(
           service_tier: Specifies the latency tier to use for processing the request. This parameter is
               relevant for customers subscribed to the scale tier service:
 
-              - If set to 'auto', the system will utilize scale tier credits until they are
-                exhausted.
+              - If set to 'auto', and the Project is Scale tier enabled, the system will
+                utilize scale tier credits until they are exhausted.
+              - If set to 'auto', and the Project is not Scale tier enabled, the request will
+                be processed using the default service tier with a lower uptime SLA and no
+                latency guarentee.
               - If set to 'default', the request will be processed using the default service
                 tier with a lower uptime SLA and no latency guarentee.
               - When not set, the default behavior is 'auto'.
@@ -1031,13 +1375,16 @@ async def create(
 
           stop: Up to 4 sequences where the API will stop generating further tokens.
 
+          store: Whether or not to store the output of this chat completion request for use in
+              our [model distillation](https://platform.openai.com/docs/guides/distillation)
+              or [evals](https://platform.openai.com/docs/guides/evals) products.
+
           stream_options: Options for streaming response. Only set this when you set `stream: true`.
 
           temperature: What sampling temperature to use, between 0 and 2. Higher values like 0.8 will
               make the output more random, while lower values like 0.2 will make it more
-              focused and deterministic.
-
-              We generally recommend altering this or `top_p` but not both.
+              focused and deterministic. We generally recommend altering this or `top_p` but
+              not both.
 
           tool_choice: Controls which (if any) tool is called by the model. `none` means the model will
               not call any tool and instead generates a message. `auto` means the model can
@@ -1065,7 +1412,7 @@ async def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -1084,19 +1431,26 @@ async def create(
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
         stream: bool,
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         response_format: completion_create_params.ResponseFormat | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
@@ -1111,15 +1465,29 @@ async def create(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> ChatCompletion | AsyncStream[ChatCompletionChunk]:
-        """
-        Creates a model response for the given chat conversation.
+        """Creates a model response for the given chat conversation.
+
+        Learn more in the
+        [text generation](https://platform.openai.com/docs/guides/text-generation),
+        [vision](https://platform.openai.com/docs/guides/vision), and
+        [audio](https://platform.openai.com/docs/guides/audio) guides.
+
+        Parameter support can differ depending on the model used to generate the
+        response, particularly for newer reasoning models. Parameters that are only
+        supported for reasoning models are noted below. For the current state of
+        unsupported parameters in reasoning models,
+        [refer to the reasoning guide](https://platform.openai.com/docs/guides/reasoning).
 
         Args:
-          messages: A list of messages comprising the conversation so far.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_format_inputs_to_chatgpt_models).
+          messages: A list of messages comprising the conversation so far. Depending on the
+              [model](https://platform.openai.com/docs/models) you use, different message
+              types (modalities) are supported, like
+              [text](https://platform.openai.com/docs/guides/text-generation),
+              [images](https://platform.openai.com/docs/guides/vision), and
+              [audio](https://platform.openai.com/docs/guides/audio).
 
           model: ID of the model to use. See the
-              [model endpoint compatibility](https://platform.openai.com/docs/models/model-endpoint-compatibility)
+              [model endpoint compatibility](https://platform.openai.com/docs/models#model-endpoint-compatibility)
               table for details on which models work with the Chat API.
 
           stream: If set, partial message deltas will be sent, like in ChatGPT. Tokens will be
@@ -1129,20 +1497,26 @@ async def create(
               message.
               [Example Python code](https://cookbook.openai.com/examples/how_to_stream_completions).
 
+          audio: Parameters for audio output. Required when audio output is requested with
+              `modalities: ["audio"]`.
+              [Learn more](https://platform.openai.com/docs/guides/audio).
+
           frequency_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on their
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
-
           function_call: Deprecated in favor of `tool_choice`.
 
-              Controls which (if any) function is called by the model. `none` means the model
-              will not call a function and instead generates a message. `auto` means the model
-              can pick between generating a message or calling a function. Specifying a
-              particular function via `{"name": "my_function"}` forces the model to call that
+              Controls which (if any) function is called by the model.
+
+              `none` means the model will not call a function and instead generates a message.
+
+              `auto` means the model can pick between generating a message or calling a
               function.
 
+              Specifying a particular function via `{"name": "my_function"}` forces the model
+              to call that function.
+
               `none` is the default when no functions are present. `auto` is the default if
               functions are present.
 
@@ -1163,35 +1537,62 @@ async def create(
               returns the log probabilities of each output token returned in the `content` of
               `message`.
 
+          max_completion_tokens: An upper bound for the number of tokens that can be generated for a completion,
+              including visible output tokens and
+              [reasoning tokens](https://platform.openai.com/docs/guides/reasoning).
+
           max_tokens: The maximum number of [tokens](/tokenizer) that can be generated in the chat
-              completion.
+              completion. This value can be used to control
+              [costs](https://openai.com/api/pricing/) for text generated via API.
+
+              This value is now deprecated in favor of `max_completion_tokens`, and is not
+              compatible with
+              [o1 series models](https://platform.openai.com/docs/guides/reasoning).
+
+          metadata: Developer-defined tags and values used for filtering completions in the
+              [dashboard](https://platform.openai.com/chat-completions).
+
+          modalities: Output types that you would like the model to generate for this request. Most
+              models are capable of generating text, which is the default:
+
+              `["text"]`
+
+              The `gpt-4o-audio-preview` model can also be used to
+              [generate audio](https://platform.openai.com/docs/guides/audio). To request that
+              this model generate both text and audio responses, you can use:
 
-              The total length of input tokens and generated tokens is limited by the model's
-              context length.
-              [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken)
-              for counting tokens.
+              `["text", "audio"]`
 
           n: How many chat completion choices to generate for each input message. Note that
               you will be charged based on the number of generated tokens across all of the
               choices. Keep `n` as `1` to minimize costs.
 
           parallel_tool_calls: Whether to enable
-              [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+              [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
               during tool use.
 
+          prediction: Static predicted output content, such as the content of a text file that is
+              being regenerated.
+
           presence_penalty: Number between -2.0 and 2.0. Positive values penalize new tokens based on
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+          reasoning_effort: **o1 models only**
+
+              Constrains effort on reasoning for
+              [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+              supported values are `low`, `medium`, and `high`. Reducing reasoning effort can
+              result in faster responses and fewer tokens used on reasoning in a response.
 
-          response_format: An object specifying the format that the model must output. Compatible with
-              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-              [GPT-4o mini](https://platform.openai.com/docs/models/gpt-4o-mini),
-              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo) and
-              all GPT-3.5 Turbo models newer than `gpt-3.5-turbo-1106`.
+          response_format: An object specifying the format that the model must output.
 
-              Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+              Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
+              [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
+
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
               message the model generates is valid JSON.
 
               **Important:** when using JSON mode, you **must** also instruct the model to
@@ -1211,8 +1612,11 @@ async def create(
           service_tier: Specifies the latency tier to use for processing the request. This parameter is
               relevant for customers subscribed to the scale tier service:
 
-              - If set to 'auto', the system will utilize scale tier credits until they are
-                exhausted.
+              - If set to 'auto', and the Project is Scale tier enabled, the system will
+                utilize scale tier credits until they are exhausted.
+              - If set to 'auto', and the Project is not Scale tier enabled, the request will
+                be processed using the default service tier with a lower uptime SLA and no
+                latency guarentee.
               - If set to 'default', the request will be processed using the default service
                 tier with a lower uptime SLA and no latency guarentee.
               - When not set, the default behavior is 'auto'.
@@ -1222,13 +1626,16 @@ async def create(
 
           stop: Up to 4 sequences where the API will stop generating further tokens.
 
+          store: Whether or not to store the output of this chat completion request for use in
+              our [model distillation](https://platform.openai.com/docs/guides/distillation)
+              or [evals](https://platform.openai.com/docs/guides/evals) products.
+
           stream_options: Options for streaming response. Only set this when you set `stream: true`.
 
           temperature: What sampling temperature to use, between 0 and 2. Higher values like 0.8 will
               make the output more random, while lower values like 0.2 will make it more
-              focused and deterministic.
-
-              We generally recommend altering this or `top_p` but not both.
+              focused and deterministic. We generally recommend altering this or `top_p` but
+              not both.
 
           tool_choice: Controls which (if any) tool is called by the model. `none` means the model will
               not call any tool and instead generates a message. `auto` means the model can
@@ -1256,7 +1663,7 @@ async def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -1274,19 +1681,26 @@ async def create(
         *,
         messages: Iterable[ChatCompletionMessageParam],
         model: Union[str, ChatModel],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
         frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
         function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
         functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
         logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
         logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
         max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | NotGiven = NOT_GIVEN,
         n: Optional[int] | NotGiven = NOT_GIVEN,
         parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
         presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: ChatCompletionReasoningEffort | NotGiven = NOT_GIVEN,
         response_format: completion_create_params.ResponseFormat | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
         stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
         stream: Optional[Literal[False]] | Literal[True] | NotGiven = NOT_GIVEN,
         stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
@@ -1302,25 +1716,33 @@ async def create(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> ChatCompletion | AsyncStream[ChatCompletionChunk]:
+        validate_response_format(response_format)
         return await self._post(
             "/chat/completions",
             body=await async_maybe_transform(
                 {
                     "messages": messages,
                     "model": model,
+                    "audio": audio,
                     "frequency_penalty": frequency_penalty,
                     "function_call": function_call,
                     "functions": functions,
                     "logit_bias": logit_bias,
                     "logprobs": logprobs,
+                    "max_completion_tokens": max_completion_tokens,
                     "max_tokens": max_tokens,
+                    "metadata": metadata,
+                    "modalities": modalities,
                     "n": n,
                     "parallel_tool_calls": parallel_tool_calls,
+                    "prediction": prediction,
                     "presence_penalty": presence_penalty,
+                    "reasoning_effort": reasoning_effort,
                     "response_format": response_format,
                     "seed": seed,
                     "service_tier": service_tier,
                     "stop": stop,
+                    "store": store,
                     "stream": stream,
                     "stream_options": stream_options,
                     "temperature": temperature,
@@ -1375,3 +1797,10 @@ def __init__(self, completions: AsyncCompletions) -> None:
         self.create = async_to_streamed_response_wrapper(
             completions.create,
         )
+
+
+def validate_response_format(response_format: object) -> None:
+    if inspect.isclass(response_format) and issubclass(response_format, pydantic.BaseModel):
+        raise TypeError(
+            "You tried to pass a `BaseModel` class to `chat.completions.create()`; You must use `beta.chat.completions.parse()` instead"
+        )
diff --git a/portkey_ai/_vendor/openai/resources/completions.py b/portkey_ai/_vendor/openai/resources/completions.py
index 0812000f..1ac3575f 100644
--- a/portkey_ai/_vendor/openai/resources/completions.py
+++ b/portkey_ai/_vendor/openai/resources/completions.py
@@ -2,8 +2,8 @@
 
 from __future__ import annotations
 
-from typing import Dict, List, Union, Iterable, Optional, overload
-from typing_extensions import Literal
+from typing import Dict, List, Union, Iterable, Optional
+from typing_extensions import Literal, overload
 
 import httpx
 
@@ -31,10 +31,21 @@
 class Completions(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> CompletionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return CompletionsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> CompletionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return CompletionsWithStreamingResponse(self)
 
     @overload
@@ -73,8 +84,8 @@ def create(
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           prompt: The prompt(s) to generate completions for, encoded as a string, array of
               strings, array of tokens, or array of token arrays.
@@ -99,7 +110,7 @@ def create(
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           logit_bias: Modify the likelihood of specified tokens appearing in the completion.
 
@@ -139,7 +150,7 @@ def create(
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           seed: If specified, our system will make a best effort to sample deterministically,
               such that repeated requests with the same `seed` and parameters should return
@@ -178,7 +189,7 @@ def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -226,8 +237,8 @@ def create(
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           prompt: The prompt(s) to generate completions for, encoded as a string, array of
               strings, array of tokens, or array of token arrays.
@@ -259,7 +270,7 @@ def create(
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           logit_bias: Modify the likelihood of specified tokens appearing in the completion.
 
@@ -299,7 +310,7 @@ def create(
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           seed: If specified, our system will make a best effort to sample deterministically,
               such that repeated requests with the same `seed` and parameters should return
@@ -331,7 +342,7 @@ def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -379,8 +390,8 @@ def create(
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           prompt: The prompt(s) to generate completions for, encoded as a string, array of
               strings, array of tokens, or array of token arrays.
@@ -412,7 +423,7 @@ def create(
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           logit_bias: Modify the likelihood of specified tokens appearing in the completion.
 
@@ -452,7 +463,7 @@ def create(
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           seed: If specified, our system will make a best effort to sample deterministically,
               such that repeated requests with the same `seed` and parameters should return
@@ -484,7 +495,7 @@ def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -562,10 +573,21 @@ def create(
 class AsyncCompletions(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncCompletionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncCompletionsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncCompletionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncCompletionsWithStreamingResponse(self)
 
     @overload
@@ -604,8 +626,8 @@ async def create(
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           prompt: The prompt(s) to generate completions for, encoded as a string, array of
               strings, array of tokens, or array of token arrays.
@@ -630,7 +652,7 @@ async def create(
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           logit_bias: Modify the likelihood of specified tokens appearing in the completion.
 
@@ -670,7 +692,7 @@ async def create(
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           seed: If specified, our system will make a best effort to sample deterministically,
               such that repeated requests with the same `seed` and parameters should return
@@ -709,7 +731,7 @@ async def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -757,8 +779,8 @@ async def create(
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           prompt: The prompt(s) to generate completions for, encoded as a string, array of
               strings, array of tokens, or array of token arrays.
@@ -790,7 +812,7 @@ async def create(
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           logit_bias: Modify the likelihood of specified tokens appearing in the completion.
 
@@ -830,7 +852,7 @@ async def create(
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           seed: If specified, our system will make a best effort to sample deterministically,
               such that repeated requests with the same `seed` and parameters should return
@@ -862,7 +884,7 @@ async def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -910,8 +932,8 @@ async def create(
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           prompt: The prompt(s) to generate completions for, encoded as a string, array of
               strings, array of tokens, or array of token arrays.
@@ -943,7 +965,7 @@ async def create(
               existing frequency in the text so far, decreasing the model's likelihood to
               repeat the same line verbatim.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           logit_bias: Modify the likelihood of specified tokens appearing in the completion.
 
@@ -983,7 +1005,7 @@ async def create(
               whether they appear in the text so far, increasing the model's likelihood to
               talk about new topics.
 
-              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+              [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
 
           seed: If specified, our system will make a best effort to sample deterministically,
               such that repeated requests with the same `seed` and parameters should return
@@ -1015,7 +1037,7 @@ async def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
diff --git a/portkey_ai/_vendor/openai/resources/embeddings.py b/portkey_ai/_vendor/openai/resources/embeddings.py
index 773b6f09..4ab2278e 100644
--- a/portkey_ai/_vendor/openai/resources/embeddings.py
+++ b/portkey_ai/_vendor/openai/resources/embeddings.py
@@ -16,9 +16,8 @@
 from .._extras import numpy as np, has_numpy
 from .._resource import SyncAPIResource, AsyncAPIResource
 from .._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
-from .._base_client import (
-    make_request_options,
-)
+from .._base_client import make_request_options
+from ..types.embedding_model import EmbeddingModel
 from ..types.create_embedding_response import CreateEmbeddingResponse
 
 __all__ = ["Embeddings", "AsyncEmbeddings"]
@@ -27,17 +26,28 @@
 class Embeddings(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> EmbeddingsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return EmbeddingsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> EmbeddingsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return EmbeddingsWithStreamingResponse(self)
 
     def create(
         self,
         *,
         input: Union[str, List[str], Iterable[int], Iterable[Iterable[int]]],
-        model: Union[str, Literal["text-embedding-ada-002", "text-embedding-3-small", "text-embedding-3-large"]],
+        model: Union[str, EmbeddingModel],
         dimensions: int | NotGiven = NOT_GIVEN,
         encoding_format: Literal["float", "base64"] | NotGiven = NOT_GIVEN,
         user: str | NotGiven = NOT_GIVEN,
@@ -63,8 +73,8 @@ def create(
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           dimensions: The number of dimensions the resulting output embeddings should have. Only
               supported in `text-embedding-3` and later models.
@@ -74,7 +84,7 @@ def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -128,17 +138,28 @@ def parser(obj: CreateEmbeddingResponse) -> CreateEmbeddingResponse:
 class AsyncEmbeddings(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncEmbeddingsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncEmbeddingsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncEmbeddingsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncEmbeddingsWithStreamingResponse(self)
 
     async def create(
         self,
         *,
         input: Union[str, List[str], Iterable[int], Iterable[Iterable[int]]],
-        model: Union[str, Literal["text-embedding-ada-002", "text-embedding-3-small", "text-embedding-3-large"]],
+        model: Union[str, EmbeddingModel],
         dimensions: int | NotGiven = NOT_GIVEN,
         encoding_format: Literal["float", "base64"] | NotGiven = NOT_GIVEN,
         user: str | NotGiven = NOT_GIVEN,
@@ -164,8 +185,8 @@ async def create(
           model: ID of the model to use. You can use the
               [List models](https://platform.openai.com/docs/api-reference/models/list) API to
               see all of your available models, or see our
-              [Model overview](https://platform.openai.com/docs/models/overview) for
-              descriptions of them.
+              [Model overview](https://platform.openai.com/docs/models) for descriptions of
+              them.
 
           dimensions: The number of dimensions the resulting output embeddings should have. Only
               supported in `text-embedding-3` and later models.
@@ -175,7 +196,7 @@ async def create(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
diff --git a/portkey_ai/_vendor/openai/resources/files.py b/portkey_ai/_vendor/openai/resources/files.py
index 75c971a8..6eaea1b5 100644
--- a/portkey_ai/_vendor/openai/resources/files.py
+++ b/portkey_ai/_vendor/openai/resources/files.py
@@ -10,7 +10,7 @@
 import httpx
 
 from .. import _legacy_response
-from ..types import file_list_params, file_create_params
+from ..types import FilePurpose, file_list_params, file_create_params
 from .._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
 from .._utils import (
     extract_files,
@@ -28,13 +28,11 @@
     to_custom_streamed_response_wrapper,
     async_to_custom_streamed_response_wrapper,
 )
-from ..pagination import SyncPage, AsyncPage
-from .._base_client import (
-    AsyncPaginator,
-    make_request_options,
-)
+from ..pagination import SyncCursorPage, AsyncCursorPage
+from .._base_client import AsyncPaginator, make_request_options
 from ..types.file_object import FileObject
 from ..types.file_deleted import FileDeleted
+from ..types.file_purpose import FilePurpose
 
 __all__ = ["Files", "AsyncFiles"]
 
@@ -42,17 +40,28 @@
 class Files(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> FilesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return FilesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> FilesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return FilesWithStreamingResponse(self)
 
     def create(
         self,
         *,
         file: FileTypes,
-        purpose: Literal["assistants", "batch", "fine-tune", "vision"],
+        purpose: FilePurpose,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -77,7 +86,7 @@ def create(
         [completions](https://platform.openai.com/docs/api-reference/fine-tuning/completions-input)
         models.
 
-        The Batch API only supports `.jsonl` files up to 100 MB in size. The input also
+        The Batch API only supports `.jsonl` files up to 200 MB in size. The input also
         has a specific required
         [format](https://platform.openai.com/docs/api-reference/batch/request-input).
 
@@ -161,6 +170,9 @@ def retrieve(
     def list(
         self,
         *,
+        after: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
         purpose: str | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -168,11 +180,23 @@ def list(
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> SyncPage[FileObject]:
-        """
-        Returns a list of files that belong to the user's organization.
+    ) -> SyncCursorPage[FileObject]:
+        """Returns a list of files.
 
         Args:
+          after: A cursor for use in pagination.
+
+        `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              10,000, and the default is 10,000.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
           purpose: Only return files with the given purpose.
 
           extra_headers: Send extra headers
@@ -185,13 +209,21 @@ def list(
         """
         return self._get_api_list(
             "/files",
-            page=SyncPage[FileObject],
+            page=SyncCursorPage[FileObject],
             options=make_request_options(
                 extra_headers=extra_headers,
                 extra_query=extra_query,
                 extra_body=extra_body,
                 timeout=timeout,
-                query=maybe_transform({"purpose": purpose}, file_list_params.FileListParams),
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "limit": limit,
+                        "order": order,
+                        "purpose": purpose,
+                    },
+                    file_list_params.FileListParams,
+                ),
             ),
             model=FileObject,
         )
@@ -324,17 +356,28 @@ def wait_for_processing(
 class AsyncFiles(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncFilesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncFilesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncFilesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncFilesWithStreamingResponse(self)
 
     async def create(
         self,
         *,
         file: FileTypes,
-        purpose: Literal["assistants", "batch", "fine-tune", "vision"],
+        purpose: FilePurpose,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -359,7 +402,7 @@ async def create(
         [completions](https://platform.openai.com/docs/api-reference/fine-tuning/completions-input)
         models.
 
-        The Batch API only supports `.jsonl` files up to 100 MB in size. The input also
+        The Batch API only supports `.jsonl` files up to 200 MB in size. The input also
         has a specific required
         [format](https://platform.openai.com/docs/api-reference/batch/request-input).
 
@@ -443,6 +486,9 @@ async def retrieve(
     def list(
         self,
         *,
+        after: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
         purpose: str | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -450,11 +496,23 @@ def list(
         extra_query: Query | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> AsyncPaginator[FileObject, AsyncPage[FileObject]]:
-        """
-        Returns a list of files that belong to the user's organization.
+    ) -> AsyncPaginator[FileObject, AsyncCursorPage[FileObject]]:
+        """Returns a list of files.
 
         Args:
+          after: A cursor for use in pagination.
+
+        `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              10,000, and the default is 10,000.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
           purpose: Only return files with the given purpose.
 
           extra_headers: Send extra headers
@@ -467,13 +525,21 @@ def list(
         """
         return self._get_api_list(
             "/files",
-            page=AsyncPage[FileObject],
+            page=AsyncCursorPage[FileObject],
             options=make_request_options(
                 extra_headers=extra_headers,
                 extra_query=extra_query,
                 extra_body=extra_body,
                 timeout=timeout,
-                query=maybe_transform({"purpose": purpose}, file_list_params.FileListParams),
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "limit": limit,
+                        "order": order,
+                        "purpose": purpose,
+                    },
+                    file_list_params.FileListParams,
+                ),
             ),
             model=FileObject,
         )
diff --git a/portkey_ai/_vendor/openai/resources/fine_tuning/fine_tuning.py b/portkey_ai/_vendor/openai/resources/fine_tuning/fine_tuning.py
index 0404fed6..d2bce87c 100644
--- a/portkey_ai/_vendor/openai/resources/fine_tuning/fine_tuning.py
+++ b/portkey_ai/_vendor/openai/resources/fine_tuning/fine_tuning.py
@@ -2,7 +2,8 @@
 
 from __future__ import annotations
 
-from .jobs import (
+from ..._compat import cached_property
+from .jobs.jobs import (
     Jobs,
     AsyncJobs,
     JobsWithRawResponse,
@@ -10,8 +11,6 @@
     JobsWithStreamingResponse,
     AsyncJobsWithStreamingResponse,
 )
-from ..._compat import cached_property
-from .jobs.jobs import Jobs, AsyncJobs
 from ..._resource import SyncAPIResource, AsyncAPIResource
 
 __all__ = ["FineTuning", "AsyncFineTuning"]
@@ -24,10 +23,21 @@ def jobs(self) -> Jobs:
 
     @cached_property
     def with_raw_response(self) -> FineTuningWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return FineTuningWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> FineTuningWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return FineTuningWithStreamingResponse(self)
 
 
@@ -38,10 +48,21 @@ def jobs(self) -> AsyncJobs:
 
     @cached_property
     def with_raw_response(self) -> AsyncFineTuningWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncFineTuningWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncFineTuningWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncFineTuningWithStreamingResponse(self)
 
 
diff --git a/portkey_ai/_vendor/openai/resources/fine_tuning/jobs/checkpoints.py b/portkey_ai/_vendor/openai/resources/fine_tuning/jobs/checkpoints.py
index 67f5739a..8b5e905e 100644
--- a/portkey_ai/_vendor/openai/resources/fine_tuning/jobs/checkpoints.py
+++ b/portkey_ai/_vendor/openai/resources/fine_tuning/jobs/checkpoints.py
@@ -24,10 +24,21 @@
 class Checkpoints(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> CheckpointsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return CheckpointsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> CheckpointsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return CheckpointsWithStreamingResponse(self)
 
     def list(
@@ -84,10 +95,21 @@ def list(
 class AsyncCheckpoints(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncCheckpointsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncCheckpointsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncCheckpointsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncCheckpointsWithStreamingResponse(self)
 
     def list(
diff --git a/portkey_ai/_vendor/openai/resources/fine_tuning/jobs/jobs.py b/portkey_ai/_vendor/openai/resources/fine_tuning/jobs/jobs.py
index 5cef7bcd..78eefc25 100644
--- a/portkey_ai/_vendor/openai/resources/fine_tuning/jobs/jobs.py
+++ b/portkey_ai/_vendor/openai/resources/fine_tuning/jobs/jobs.py
@@ -43,10 +43,21 @@ def checkpoints(self) -> Checkpoints:
 
     @cached_property
     def with_raw_response(self) -> JobsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return JobsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> JobsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return JobsWithStreamingResponse(self)
 
     def create(
@@ -56,6 +67,7 @@ def create(
         training_file: str,
         hyperparameters: job_create_params.Hyperparameters | NotGiven = NOT_GIVEN,
         integrations: Optional[Iterable[job_create_params.Integration]] | NotGiven = NOT_GIVEN,
+        method: job_create_params.Method | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         suffix: Optional[str] | NotGiven = NOT_GIVEN,
         validation_file: Optional[str] | NotGiven = NOT_GIVEN,
@@ -77,7 +89,7 @@ def create(
 
         Args:
           model: The name of the model to fine-tune. You can select one of the
-              [supported models](https://platform.openai.com/docs/guides/fine-tuning/which-models-can-be-fine-tuned).
+              [supported models](https://platform.openai.com/docs/guides/fine-tuning#which-models-can-be-fine-tuned).
 
           training_file: The ID of an uploaded file that contains training data.
 
@@ -88,22 +100,27 @@ def create(
               your file with the purpose `fine-tune`.
 
               The contents of the file should differ depending on if the model uses the
-              [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input) or
+              [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input),
               [completions](https://platform.openai.com/docs/api-reference/fine-tuning/completions-input)
+              format, or if the fine-tuning method uses the
+              [preference](https://platform.openai.com/docs/api-reference/fine-tuning/preference-input)
               format.
 
               See the [fine-tuning guide](https://platform.openai.com/docs/guides/fine-tuning)
               for more details.
 
-          hyperparameters: The hyperparameters used for the fine-tuning job.
+          hyperparameters: The hyperparameters used for the fine-tuning job. This value is now deprecated
+              in favor of `method`, and should be passed in under the `method` parameter.
 
           integrations: A list of integrations to enable for your fine-tuning job.
 
+          method: The method used for fine-tuning.
+
           seed: The seed controls the reproducibility of the job. Passing in the same seed and
               job parameters should produce the same results, but may differ in rare cases. If
               a seed is not specified, one will be generated for you.
 
-          suffix: A string of up to 18 characters that will be added to your fine-tuned model
+          suffix: A string of up to 64 characters that will be added to your fine-tuned model
               name.
 
               For example, a `suffix` of "custom-model-name" would produce a model name like
@@ -138,6 +155,7 @@ def create(
                     "training_file": training_file,
                     "hyperparameters": hyperparameters,
                     "integrations": integrations,
+                    "method": method,
                     "seed": seed,
                     "suffix": suffix,
                     "validation_file": validation_file,
@@ -323,10 +341,21 @@ def checkpoints(self) -> AsyncCheckpoints:
 
     @cached_property
     def with_raw_response(self) -> AsyncJobsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncJobsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncJobsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncJobsWithStreamingResponse(self)
 
     async def create(
@@ -336,6 +365,7 @@ async def create(
         training_file: str,
         hyperparameters: job_create_params.Hyperparameters | NotGiven = NOT_GIVEN,
         integrations: Optional[Iterable[job_create_params.Integration]] | NotGiven = NOT_GIVEN,
+        method: job_create_params.Method | NotGiven = NOT_GIVEN,
         seed: Optional[int] | NotGiven = NOT_GIVEN,
         suffix: Optional[str] | NotGiven = NOT_GIVEN,
         validation_file: Optional[str] | NotGiven = NOT_GIVEN,
@@ -357,7 +387,7 @@ async def create(
 
         Args:
           model: The name of the model to fine-tune. You can select one of the
-              [supported models](https://platform.openai.com/docs/guides/fine-tuning/which-models-can-be-fine-tuned).
+              [supported models](https://platform.openai.com/docs/guides/fine-tuning#which-models-can-be-fine-tuned).
 
           training_file: The ID of an uploaded file that contains training data.
 
@@ -368,22 +398,27 @@ async def create(
               your file with the purpose `fine-tune`.
 
               The contents of the file should differ depending on if the model uses the
-              [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input) or
+              [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input),
               [completions](https://platform.openai.com/docs/api-reference/fine-tuning/completions-input)
+              format, or if the fine-tuning method uses the
+              [preference](https://platform.openai.com/docs/api-reference/fine-tuning/preference-input)
               format.
 
               See the [fine-tuning guide](https://platform.openai.com/docs/guides/fine-tuning)
               for more details.
 
-          hyperparameters: The hyperparameters used for the fine-tuning job.
+          hyperparameters: The hyperparameters used for the fine-tuning job. This value is now deprecated
+              in favor of `method`, and should be passed in under the `method` parameter.
 
           integrations: A list of integrations to enable for your fine-tuning job.
 
+          method: The method used for fine-tuning.
+
           seed: The seed controls the reproducibility of the job. Passing in the same seed and
               job parameters should produce the same results, but may differ in rare cases. If
               a seed is not specified, one will be generated for you.
 
-          suffix: A string of up to 18 characters that will be added to your fine-tuned model
+          suffix: A string of up to 64 characters that will be added to your fine-tuned model
               name.
 
               For example, a `suffix` of "custom-model-name" would produce a model name like
@@ -418,6 +453,7 @@ async def create(
                     "training_file": training_file,
                     "hyperparameters": hyperparameters,
                     "integrations": integrations,
+                    "method": method,
                     "seed": seed,
                     "suffix": suffix,
                     "validation_file": validation_file,
diff --git a/portkey_ai/_vendor/openai/resources/images.py b/portkey_ai/_vendor/openai/resources/images.py
index 0913b572..2fbc077d 100644
--- a/portkey_ai/_vendor/openai/resources/images.py
+++ b/portkey_ai/_vendor/openai/resources/images.py
@@ -29,10 +29,21 @@
 class Images(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> ImagesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return ImagesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> ImagesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return ImagesWithStreamingResponse(self)
 
     def create_variation(
@@ -73,7 +84,7 @@ def create_variation(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -154,7 +165,7 @@ def edit(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -240,7 +251,7 @@ def generate(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -275,10 +286,21 @@ def generate(
 class AsyncImages(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncImagesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncImagesWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncImagesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncImagesWithStreamingResponse(self)
 
     async def create_variation(
@@ -319,7 +341,7 @@ async def create_variation(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -400,7 +422,7 @@ async def edit(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
@@ -486,7 +508,7 @@ async def generate(
 
           user: A unique identifier representing your end-user, which can help OpenAI to monitor
               and detect abuse.
-              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
 
           extra_headers: Send extra headers
 
diff --git a/portkey_ai/_vendor/openai/resources/models.py b/portkey_ai/_vendor/openai/resources/models.py
index e76c496f..d6062de2 100644
--- a/portkey_ai/_vendor/openai/resources/models.py
+++ b/portkey_ai/_vendor/openai/resources/models.py
@@ -23,10 +23,21 @@
 class Models(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> ModelsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return ModelsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> ModelsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return ModelsWithStreamingResponse(self)
 
     def retrieve(
@@ -125,10 +136,21 @@ def delete(
 class AsyncModels(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncModelsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncModelsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncModelsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncModelsWithStreamingResponse(self)
 
     async def retrieve(
diff --git a/portkey_ai/_vendor/openai/resources/moderations.py b/portkey_ai/_vendor/openai/resources/moderations.py
index b9ad9972..ce80bb7d 100644
--- a/portkey_ai/_vendor/openai/resources/moderations.py
+++ b/portkey_ai/_vendor/openai/resources/moderations.py
@@ -2,7 +2,7 @@
 
 from __future__ import annotations
 
-from typing import List, Union
+from typing import List, Union, Iterable
 
 import httpx
 
@@ -19,6 +19,7 @@
 from .._base_client import make_request_options
 from ..types.moderation_model import ModerationModel
 from ..types.moderation_create_response import ModerationCreateResponse
+from ..types.moderation_multi_modal_input_param import ModerationMultiModalInputParam
 
 __all__ = ["Moderations", "AsyncModerations"]
 
@@ -26,16 +27,27 @@
 class Moderations(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> ModerationsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return ModerationsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> ModerationsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return ModerationsWithStreamingResponse(self)
 
     def create(
         self,
         *,
-        input: Union[str, List[str]],
+        input: Union[str, List[str], Iterable[ModerationMultiModalInputParam]],
         model: Union[str, ModerationModel] | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -44,20 +56,19 @@ def create(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> ModerationCreateResponse:
-        """
-        Classifies if text is potentially harmful.
+        """Classifies if text and/or image inputs are potentially harmful.
 
-        Args:
-          input: The input text to classify
+        Learn more in
+        the [moderation guide](https://platform.openai.com/docs/guides/moderation).
 
-          model: Two content moderations models are available: `text-moderation-stable` and
-              `text-moderation-latest`.
+        Args:
+          input: Input (or inputs) to classify. Can be a single string, an array of strings, or
+              an array of multi-modal input objects similar to other models.
 
-              The default is `text-moderation-latest` which will be automatically upgraded
-              over time. This ensures you are always using our most accurate model. If you use
-              `text-moderation-stable`, we will provide advanced notice before updating the
-              model. Accuracy of `text-moderation-stable` may be slightly lower than for
-              `text-moderation-latest`.
+          model: The content moderation model you would like to use. Learn more in
+              [the moderation guide](https://platform.openai.com/docs/guides/moderation), and
+              learn about available models
+              [here](https://platform.openai.com/docs/models#moderation).
 
           extra_headers: Send extra headers
 
@@ -86,16 +97,27 @@ def create(
 class AsyncModerations(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncModerationsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncModerationsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncModerationsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncModerationsWithStreamingResponse(self)
 
     async def create(
         self,
         *,
-        input: Union[str, List[str]],
+        input: Union[str, List[str], Iterable[ModerationMultiModalInputParam]],
         model: Union[str, ModerationModel] | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -104,20 +126,19 @@ async def create(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> ModerationCreateResponse:
-        """
-        Classifies if text is potentially harmful.
+        """Classifies if text and/or image inputs are potentially harmful.
 
-        Args:
-          input: The input text to classify
+        Learn more in
+        the [moderation guide](https://platform.openai.com/docs/guides/moderation).
 
-          model: Two content moderations models are available: `text-moderation-stable` and
-              `text-moderation-latest`.
+        Args:
+          input: Input (or inputs) to classify. Can be a single string, an array of strings, or
+              an array of multi-modal input objects similar to other models.
 
-              The default is `text-moderation-latest` which will be automatically upgraded
-              over time. This ensures you are always using our most accurate model. If you use
-              `text-moderation-stable`, we will provide advanced notice before updating the
-              model. Accuracy of `text-moderation-stable` may be slightly lower than for
-              `text-moderation-latest`.
+          model: The content moderation model you would like to use. Learn more in
+              [the moderation guide](https://platform.openai.com/docs/guides/moderation), and
+              learn about available models
+              [here](https://platform.openai.com/docs/models#moderation).
 
           extra_headers: Send extra headers
 
diff --git a/portkey_ai/_vendor/openai/resources/uploads/parts.py b/portkey_ai/_vendor/openai/resources/uploads/parts.py
index 3ec2592b..d46e5ea1 100644
--- a/portkey_ai/_vendor/openai/resources/uploads/parts.py
+++ b/portkey_ai/_vendor/openai/resources/uploads/parts.py
@@ -27,10 +27,21 @@
 class Parts(SyncAPIResource):
     @cached_property
     def with_raw_response(self) -> PartsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return PartsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> PartsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return PartsWithStreamingResponse(self)
 
     def create(
@@ -91,10 +102,21 @@ def create(
 class AsyncParts(AsyncAPIResource):
     @cached_property
     def with_raw_response(self) -> AsyncPartsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncPartsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncPartsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncPartsWithStreamingResponse(self)
 
     async def create(
diff --git a/portkey_ai/_vendor/openai/resources/uploads/uploads.py b/portkey_ai/_vendor/openai/resources/uploads/uploads.py
index 4100423d..cfb500b6 100644
--- a/portkey_ai/_vendor/openai/resources/uploads/uploads.py
+++ b/portkey_ai/_vendor/openai/resources/uploads/uploads.py
@@ -2,9 +2,14 @@
 
 from __future__ import annotations
 
-from typing import List
-from typing_extensions import Literal
-
+import io
+import os
+import logging
+import builtins
+from typing import List, overload
+from pathlib import Path
+
+import anyio
 import httpx
 
 from ... import _legacy_response
@@ -16,7 +21,7 @@
     PartsWithStreamingResponse,
     AsyncPartsWithStreamingResponse,
 )
-from ...types import upload_create_params, upload_complete_params
+from ...types import FilePurpose, upload_create_params, upload_complete_params
 from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven
 from ..._utils import (
     maybe_transform,
@@ -27,10 +32,17 @@
 from ..._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
 from ..._base_client import make_request_options
 from ...types.upload import Upload
+from ...types.file_purpose import FilePurpose
 
 __all__ = ["Uploads", "AsyncUploads"]
 
 
+# 64MB
+DEFAULT_PART_SIZE = 64 * 1024 * 1024
+
+log: logging.Logger = logging.getLogger(__name__)
+
+
 class Uploads(SyncAPIResource):
     @cached_property
     def parts(self) -> Parts:
@@ -38,19 +50,129 @@ def parts(self) -> Parts:
 
     @cached_property
     def with_raw_response(self) -> UploadsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return UploadsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> UploadsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return UploadsWithStreamingResponse(self)
 
+    @overload
+    def upload_file_chunked(
+        self,
+        *,
+        file: os.PathLike[str],
+        mime_type: str,
+        purpose: FilePurpose,
+        bytes: int | None = None,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits a file into multiple 64MB parts and uploads them sequentially."""
+
+    @overload
+    def upload_file_chunked(
+        self,
+        *,
+        file: bytes,
+        filename: str,
+        bytes: int,
+        mime_type: str,
+        purpose: FilePurpose,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits an in-memory file into multiple 64MB parts and uploads them sequentially."""
+
+    def upload_file_chunked(
+        self,
+        *,
+        file: os.PathLike[str] | bytes,
+        mime_type: str,
+        purpose: FilePurpose,
+        filename: str | None = None,
+        bytes: int | None = None,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits the given file into multiple parts and uploads them sequentially.
+
+        ```py
+        from pathlib import Path
+
+        client.uploads.upload_file(
+            file=Path("my-paper.pdf"),
+            mime_type="pdf",
+            purpose="assistants",
+        )
+        ```
+        """
+        if isinstance(file, builtins.bytes):
+            if filename is None:
+                raise TypeError("The `filename` argument must be given for in-memory files")
+
+            if bytes is None:
+                raise TypeError("The `bytes` argument must be given for in-memory files")
+        else:
+            if not isinstance(file, Path):
+                file = Path(file)
+
+            if not filename:
+                filename = file.name
+
+            if bytes is None:
+                bytes = file.stat().st_size
+
+        upload = self.create(
+            bytes=bytes,
+            filename=filename,
+            mime_type=mime_type,
+            purpose=purpose,
+        )
+
+        part_ids: list[str] = []
+
+        if part_size is None:
+            part_size = DEFAULT_PART_SIZE
+
+        if isinstance(file, builtins.bytes):
+            buf: io.FileIO | io.BytesIO = io.BytesIO(file)
+        else:
+            buf = io.FileIO(file)
+
+        try:
+            while True:
+                data = buf.read(part_size)
+                if not data:
+                    # EOF
+                    break
+
+                part = self.parts.create(upload_id=upload.id, data=data)
+                log.info("Uploaded part %s for upload %s", part.id, upload.id)
+                part_ids.append(part.id)
+        except Exception:
+            buf.close()
+            raise
+
+        return self.complete(upload_id=upload.id, part_ids=part_ids, md5=md5)
+
     def create(
         self,
         *,
         bytes: int,
         filename: str,
         mime_type: str,
-        purpose: Literal["assistants", "batch", "fine-tune", "vision"],
+        purpose: FilePurpose,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -74,7 +196,7 @@ def create(
         For certain `purpose`s, the correct `mime_type` must be specified. Please refer
         to documentation for the supported MIME types for your use case:
 
-        - [Assistants](https://platform.openai.com/docs/assistants/tools/file-search/supported-files)
+        - [Assistants](https://platform.openai.com/docs/assistants/tools/file-search#supported-files)
 
         For guidance on the proper filename extensions for each purpose, please follow
         the documentation on
@@ -221,19 +343,140 @@ def parts(self) -> AsyncParts:
 
     @cached_property
     def with_raw_response(self) -> AsyncUploadsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
         return AsyncUploadsWithRawResponse(self)
 
     @cached_property
     def with_streaming_response(self) -> AsyncUploadsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
         return AsyncUploadsWithStreamingResponse(self)
 
+    @overload
+    async def upload_file_chunked(
+        self,
+        *,
+        file: os.PathLike[str],
+        mime_type: str,
+        purpose: FilePurpose,
+        bytes: int | None = None,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits a file into multiple 64MB parts and uploads them sequentially."""
+
+    @overload
+    async def upload_file_chunked(
+        self,
+        *,
+        file: bytes,
+        filename: str,
+        bytes: int,
+        mime_type: str,
+        purpose: FilePurpose,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits an in-memory file into multiple 64MB parts and uploads them sequentially."""
+
+    async def upload_file_chunked(
+        self,
+        *,
+        file: os.PathLike[str] | bytes,
+        mime_type: str,
+        purpose: FilePurpose,
+        filename: str | None = None,
+        bytes: int | None = None,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits the given file into multiple parts and uploads them sequentially.
+
+        ```py
+        from pathlib import Path
+
+        client.uploads.upload_file(
+            file=Path("my-paper.pdf"),
+            mime_type="pdf",
+            purpose="assistants",
+        )
+        ```
+        """
+        if isinstance(file, builtins.bytes):
+            if filename is None:
+                raise TypeError("The `filename` argument must be given for in-memory files")
+
+            if bytes is None:
+                raise TypeError("The `bytes` argument must be given for in-memory files")
+        else:
+            if not isinstance(file, anyio.Path):
+                file = anyio.Path(file)
+
+            if not filename:
+                filename = file.name
+
+            if bytes is None:
+                stat = await file.stat()
+                bytes = stat.st_size
+
+        upload = await self.create(
+            bytes=bytes,
+            filename=filename,
+            mime_type=mime_type,
+            purpose=purpose,
+        )
+
+        part_ids: list[str] = []
+
+        if part_size is None:
+            part_size = DEFAULT_PART_SIZE
+
+        if isinstance(file, anyio.Path):
+            fd = await file.open("rb")
+            async with fd:
+                while True:
+                    data = await fd.read(part_size)
+                    if not data:
+                        # EOF
+                        break
+
+                    part = await self.parts.create(upload_id=upload.id, data=data)
+                    log.info("Uploaded part %s for upload %s", part.id, upload.id)
+                    part_ids.append(part.id)
+        else:
+            buf = io.BytesIO(file)
+
+            try:
+                while True:
+                    data = buf.read(part_size)
+                    if not data:
+                        # EOF
+                        break
+
+                    part = await self.parts.create(upload_id=upload.id, data=data)
+                    log.info("Uploaded part %s for upload %s", part.id, upload.id)
+                    part_ids.append(part.id)
+            except Exception:
+                buf.close()
+                raise
+
+        return await self.complete(upload_id=upload.id, part_ids=part_ids, md5=md5)
+
     async def create(
         self,
         *,
         bytes: int,
         filename: str,
         mime_type: str,
-        purpose: Literal["assistants", "batch", "fine-tune", "vision"],
+        purpose: FilePurpose,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
         extra_headers: Headers | None = None,
@@ -257,7 +500,7 @@ async def create(
         For certain `purpose`s, the correct `mime_type` must be specified. Please refer
         to documentation for the supported MIME types for your use case:
 
-        - [Assistants](https://platform.openai.com/docs/assistants/tools/file-search/supported-files)
+        - [Assistants](https://platform.openai.com/docs/assistants/tools/file-search#supported-files)
 
         For guidance on the proper filename extensions for each purpose, please follow
         the documentation on
diff --git a/portkey_ai/_vendor/openai/types/__init__.py b/portkey_ai/_vendor/openai/types/__init__.py
index f621fb67..72950f24 100644
--- a/portkey_ai/_vendor/openai/types/__init__.py
+++ b/portkey_ai/_vendor/openai/types/__init__.py
@@ -24,7 +24,9 @@
 from .image_model import ImageModel as ImageModel
 from .file_content import FileContent as FileContent
 from .file_deleted import FileDeleted as FileDeleted
+from .file_purpose import FilePurpose as FilePurpose
 from .model_deleted import ModelDeleted as ModelDeleted
+from .embedding_model import EmbeddingModel as EmbeddingModel
 from .images_response import ImagesResponse as ImagesResponse
 from .completion_usage import CompletionUsage as CompletionUsage
 from .file_list_params import FileListParams as FileListParams
@@ -36,6 +38,7 @@
 from .batch_create_params import BatchCreateParams as BatchCreateParams
 from .batch_request_counts import BatchRequestCounts as BatchRequestCounts
 from .upload_create_params import UploadCreateParams as UploadCreateParams
+from .audio_response_format import AudioResponseFormat as AudioResponseFormat
 from .image_generate_params import ImageGenerateParams as ImageGenerateParams
 from .upload_complete_params import UploadCompleteParams as UploadCompleteParams
 from .embedding_create_params import EmbeddingCreateParams as EmbeddingCreateParams
@@ -43,4 +46,8 @@
 from .moderation_create_params import ModerationCreateParams as ModerationCreateParams
 from .create_embedding_response import CreateEmbeddingResponse as CreateEmbeddingResponse
 from .moderation_create_response import ModerationCreateResponse as ModerationCreateResponse
+from .moderation_text_input_param import ModerationTextInputParam as ModerationTextInputParam
+from .websocket_connection_options import WebsocketConnectionOptions as WebsocketConnectionOptions
 from .image_create_variation_params import ImageCreateVariationParams as ImageCreateVariationParams
+from .moderation_image_url_input_param import ModerationImageURLInputParam as ModerationImageURLInputParam
+from .moderation_multi_modal_input_param import ModerationMultiModalInputParam as ModerationMultiModalInputParam
diff --git a/portkey_ai/_vendor/openai/types/audio/__init__.py b/portkey_ai/_vendor/openai/types/audio/__init__.py
index 1de5c0ff..822e0f3a 100644
--- a/portkey_ai/_vendor/openai/types/audio/__init__.py
+++ b/portkey_ai/_vendor/openai/types/audio/__init__.py
@@ -5,6 +5,12 @@
 from .translation import Translation as Translation
 from .speech_model import SpeechModel as SpeechModel
 from .transcription import Transcription as Transcription
+from .transcription_word import TranscriptionWord as TranscriptionWord
+from .translation_verbose import TranslationVerbose as TranslationVerbose
 from .speech_create_params import SpeechCreateParams as SpeechCreateParams
+from .transcription_segment import TranscriptionSegment as TranscriptionSegment
+from .transcription_verbose import TranscriptionVerbose as TranscriptionVerbose
 from .translation_create_params import TranslationCreateParams as TranslationCreateParams
 from .transcription_create_params import TranscriptionCreateParams as TranscriptionCreateParams
+from .translation_create_response import TranslationCreateResponse as TranslationCreateResponse
+from .transcription_create_response import TranscriptionCreateResponse as TranscriptionCreateResponse
diff --git a/portkey_ai/_vendor/openai/types/audio/speech_create_params.py b/portkey_ai/_vendor/openai/types/audio/speech_create_params.py
index dff66e49..a60d0007 100644
--- a/portkey_ai/_vendor/openai/types/audio/speech_create_params.py
+++ b/portkey_ai/_vendor/openai/types/audio/speech_create_params.py
@@ -16,7 +16,7 @@ class SpeechCreateParams(TypedDict, total=False):
 
     model: Required[Union[str, SpeechModel]]
     """
-    One of the available [TTS models](https://platform.openai.com/docs/models/tts):
+    One of the available [TTS models](https://platform.openai.com/docs/models#tts):
     `tts-1` or `tts-1-hd`
     """
 
@@ -25,7 +25,7 @@ class SpeechCreateParams(TypedDict, total=False):
 
     Supported voices are `alloy`, `echo`, `fable`, `onyx`, `nova`, and `shimmer`.
     Previews of the voices are available in the
-    [Text to speech guide](https://platform.openai.com/docs/guides/text-to-speech/voice-options).
+    [Text to speech guide](https://platform.openai.com/docs/guides/text-to-speech#voice-options).
     """
 
     response_format: Literal["mp3", "opus", "aac", "flac", "wav", "pcm"]
diff --git a/portkey_ai/_vendor/openai/types/audio/transcription_create_params.py b/portkey_ai/_vendor/openai/types/audio/transcription_create_params.py
index a825fefe..88805aff 100644
--- a/portkey_ai/_vendor/openai/types/audio/transcription_create_params.py
+++ b/portkey_ai/_vendor/openai/types/audio/transcription_create_params.py
@@ -7,6 +7,7 @@
 
 from ..._types import FileTypes
 from ..audio_model import AudioModel
+from ..audio_response_format import AudioResponseFormat
 
 __all__ = ["TranscriptionCreateParams"]
 
@@ -37,14 +38,14 @@ class TranscriptionCreateParams(TypedDict, total=False):
     """An optional text to guide the model's style or continue a previous audio
     segment.
 
-    The [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
+    The [prompt](https://platform.openai.com/docs/guides/speech-to-text#prompting)
     should match the audio language.
     """
 
-    response_format: Literal["json", "text", "srt", "verbose_json", "vtt"]
+    response_format: AudioResponseFormat
     """
-    The format of the transcript output, in one of these options: `json`, `text`,
-    `srt`, `verbose_json`, or `vtt`.
+    The format of the output, in one of these options: `json`, `text`, `srt`,
+    `verbose_json`, or `vtt`.
     """
 
     temperature: float
diff --git a/portkey_ai/_vendor/openai/types/audio/transcription_create_response.py b/portkey_ai/_vendor/openai/types/audio/transcription_create_response.py
new file mode 100644
index 00000000..2f7bed81
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/audio/transcription_create_response.py
@@ -0,0 +1,11 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Union
+from typing_extensions import TypeAlias
+
+from .transcription import Transcription
+from .transcription_verbose import TranscriptionVerbose
+
+__all__ = ["TranscriptionCreateResponse"]
+
+TranscriptionCreateResponse: TypeAlias = Union[Transcription, TranscriptionVerbose]
diff --git a/portkey_ai/_vendor/openai/types/audio/transcription_segment.py b/portkey_ai/_vendor/openai/types/audio/transcription_segment.py
new file mode 100644
index 00000000..522c401e
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/audio/transcription_segment.py
@@ -0,0 +1,49 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List
+
+from ..._models import BaseModel
+
+__all__ = ["TranscriptionSegment"]
+
+
+class TranscriptionSegment(BaseModel):
+    id: int
+    """Unique identifier of the segment."""
+
+    avg_logprob: float
+    """Average logprob of the segment.
+
+    If the value is lower than -1, consider the logprobs failed.
+    """
+
+    compression_ratio: float
+    """Compression ratio of the segment.
+
+    If the value is greater than 2.4, consider the compression failed.
+    """
+
+    end: float
+    """End time of the segment in seconds."""
+
+    no_speech_prob: float
+    """Probability of no speech in the segment.
+
+    If the value is higher than 1.0 and the `avg_logprob` is below -1, consider this
+    segment silent.
+    """
+
+    seek: int
+    """Seek offset of the segment."""
+
+    start: float
+    """Start time of the segment in seconds."""
+
+    temperature: float
+    """Temperature parameter used for generating the segment."""
+
+    text: str
+    """Text content of the segment."""
+
+    tokens: List[int]
+    """Array of token IDs for the text content."""
diff --git a/portkey_ai/_vendor/openai/types/audio/transcription_verbose.py b/portkey_ai/_vendor/openai/types/audio/transcription_verbose.py
new file mode 100644
index 00000000..3b18fa48
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/audio/transcription_verbose.py
@@ -0,0 +1,26 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List, Optional
+
+from ..._models import BaseModel
+from .transcription_word import TranscriptionWord
+from .transcription_segment import TranscriptionSegment
+
+__all__ = ["TranscriptionVerbose"]
+
+
+class TranscriptionVerbose(BaseModel):
+    duration: str
+    """The duration of the input audio."""
+
+    language: str
+    """The language of the input audio."""
+
+    text: str
+    """The transcribed text."""
+
+    segments: Optional[List[TranscriptionSegment]] = None
+    """Segments of the transcribed text and their corresponding details."""
+
+    words: Optional[List[TranscriptionWord]] = None
+    """Extracted words and their corresponding timestamps."""
diff --git a/portkey_ai/_vendor/openai/types/audio/transcription_word.py b/portkey_ai/_vendor/openai/types/audio/transcription_word.py
new file mode 100644
index 00000000..969da325
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/audio/transcription_word.py
@@ -0,0 +1,17 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+
+from ..._models import BaseModel
+
+__all__ = ["TranscriptionWord"]
+
+
+class TranscriptionWord(BaseModel):
+    end: float
+    """End time of the word in seconds."""
+
+    start: float
+    """Start time of the word in seconds."""
+
+    word: str
+    """The text content of the word."""
diff --git a/portkey_ai/_vendor/openai/types/audio/translation_create_params.py b/portkey_ai/_vendor/openai/types/audio/translation_create_params.py
index 054996a1..62f85b87 100644
--- a/portkey_ai/_vendor/openai/types/audio/translation_create_params.py
+++ b/portkey_ai/_vendor/openai/types/audio/translation_create_params.py
@@ -7,6 +7,7 @@
 
 from ..._types import FileTypes
 from ..audio_model import AudioModel
+from ..audio_response_format import AudioResponseFormat
 
 __all__ = ["TranslationCreateParams"]
 
@@ -29,14 +30,14 @@ class TranslationCreateParams(TypedDict, total=False):
     """An optional text to guide the model's style or continue a previous audio
     segment.
 
-    The [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
+    The [prompt](https://platform.openai.com/docs/guides/speech-to-text#prompting)
     should be in English.
     """
 
-    response_format: str
+    response_format: AudioResponseFormat
     """
-    The format of the transcript output, in one of these options: `json`, `text`,
-    `srt`, `verbose_json`, or `vtt`.
+    The format of the output, in one of these options: `json`, `text`, `srt`,
+    `verbose_json`, or `vtt`.
     """
 
     temperature: float
diff --git a/portkey_ai/_vendor/openai/types/audio/translation_create_response.py b/portkey_ai/_vendor/openai/types/audio/translation_create_response.py
new file mode 100644
index 00000000..9953813c
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/audio/translation_create_response.py
@@ -0,0 +1,11 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Union
+from typing_extensions import TypeAlias
+
+from .translation import Translation
+from .translation_verbose import TranslationVerbose
+
+__all__ = ["TranslationCreateResponse"]
+
+TranslationCreateResponse: TypeAlias = Union[Translation, TranslationVerbose]
diff --git a/portkey_ai/_vendor/openai/types/audio/translation_verbose.py b/portkey_ai/_vendor/openai/types/audio/translation_verbose.py
new file mode 100644
index 00000000..5901ae75
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/audio/translation_verbose.py
@@ -0,0 +1,22 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List, Optional
+
+from ..._models import BaseModel
+from .transcription_segment import TranscriptionSegment
+
+__all__ = ["TranslationVerbose"]
+
+
+class TranslationVerbose(BaseModel):
+    duration: str
+    """The duration of the input audio."""
+
+    language: str
+    """The language of the output translation (always `english`)."""
+
+    text: str
+    """The translated text."""
+
+    segments: Optional[List[TranscriptionSegment]] = None
+    """Segments of the translated text and their corresponding details."""
diff --git a/portkey_ai/_vendor/openai/types/audio_response_format.py b/portkey_ai/_vendor/openai/types/audio_response_format.py
new file mode 100644
index 00000000..f8c8d459
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/audio_response_format.py
@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal, TypeAlias
+
+__all__ = ["AudioResponseFormat"]
+
+AudioResponseFormat: TypeAlias = Literal["json", "text", "srt", "verbose_json", "vtt"]
diff --git a/portkey_ai/_vendor/openai/types/batch.py b/portkey_ai/_vendor/openai/types/batch.py
index 90f6d795..ac3d7ea1 100644
--- a/portkey_ai/_vendor/openai/types/batch.py
+++ b/portkey_ai/_vendor/openai/types/batch.py
@@ -75,7 +75,7 @@ class Batch(BaseModel):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     output_file_id: Optional[str] = None
diff --git a/portkey_ai/_vendor/openai/types/batch_create_params.py b/portkey_ai/_vendor/openai/types/batch_create_params.py
index 55517d28..b30c4d46 100644
--- a/portkey_ai/_vendor/openai/types/batch_create_params.py
+++ b/portkey_ai/_vendor/openai/types/batch_create_params.py
@@ -32,7 +32,7 @@ class BatchCreateParams(TypedDict, total=False):
     Your input file must be formatted as a
     [JSONL file](https://platform.openai.com/docs/api-reference/batch/request-input),
     and must be uploaded with the purpose `batch`. The file can contain up to 50,000
-    requests, and can be up to 100 MB in size.
+    requests, and can be up to 200 MB in size.
     """
 
     metadata: Optional[Dict[str, str]]
diff --git a/portkey_ai/_vendor/openai/types/beta/__init__.py b/portkey_ai/_vendor/openai/types/beta/__init__.py
index 9c5ddfdb..7f76fed0 100644
--- a/portkey_ai/_vendor/openai/types/beta/__init__.py
+++ b/portkey_ai/_vendor/openai/types/beta/__init__.py
@@ -19,6 +19,7 @@
 from .assistant_tool_choice import AssistantToolChoice as AssistantToolChoice
 from .code_interpreter_tool import CodeInterpreterTool as CodeInterpreterTool
 from .assistant_stream_event import AssistantStreamEvent as AssistantStreamEvent
+from .file_chunking_strategy import FileChunkingStrategy as FileChunkingStrategy
 from .file_search_tool_param import FileSearchToolParam as FileSearchToolParam
 from .assistant_create_params import AssistantCreateParams as AssistantCreateParams
 from .assistant_update_params import AssistantUpdateParams as AssistantUpdateParams
@@ -28,11 +29,17 @@
 from .assistant_tool_choice_param import AssistantToolChoiceParam as AssistantToolChoiceParam
 from .code_interpreter_tool_param import CodeInterpreterToolParam as CodeInterpreterToolParam
 from .assistant_tool_choice_option import AssistantToolChoiceOption as AssistantToolChoiceOption
+from .file_chunking_strategy_param import FileChunkingStrategyParam as FileChunkingStrategyParam
 from .thread_create_and_run_params import ThreadCreateAndRunParams as ThreadCreateAndRunParams
+from .static_file_chunking_strategy import StaticFileChunkingStrategy as StaticFileChunkingStrategy
 from .assistant_tool_choice_function import AssistantToolChoiceFunction as AssistantToolChoiceFunction
 from .assistant_response_format_option import AssistantResponseFormatOption as AssistantResponseFormatOption
+from .auto_file_chunking_strategy_param import AutoFileChunkingStrategyParam as AutoFileChunkingStrategyParam
 from .assistant_tool_choice_option_param import AssistantToolChoiceOptionParam as AssistantToolChoiceOptionParam
+from .other_file_chunking_strategy_object import OtherFileChunkingStrategyObject as OtherFileChunkingStrategyObject
+from .static_file_chunking_strategy_param import StaticFileChunkingStrategyParam as StaticFileChunkingStrategyParam
 from .assistant_tool_choice_function_param import AssistantToolChoiceFunctionParam as AssistantToolChoiceFunctionParam
+from .static_file_chunking_strategy_object import StaticFileChunkingStrategyObject as StaticFileChunkingStrategyObject
 from .assistant_response_format_option_param import (
     AssistantResponseFormatOptionParam as AssistantResponseFormatOptionParam,
 )
diff --git a/portkey_ai/_vendor/openai/types/beta/assistant.py b/portkey_ai/_vendor/openai/types/beta/assistant.py
index c6a0a4cf..3c8b8e40 100644
--- a/portkey_ai/_vendor/openai/types/beta/assistant.py
+++ b/portkey_ai/_vendor/openai/types/beta/assistant.py
@@ -56,7 +56,7 @@ class Assistant(BaseModel):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     model: str
@@ -65,8 +65,8 @@ class Assistant(BaseModel):
     You can use the
     [List models](https://platform.openai.com/docs/api-reference/models/list) API to
     see all of your available models, or see our
-    [Model overview](https://platform.openai.com/docs/models/overview) for
-    descriptions of them.
+    [Model overview](https://platform.openai.com/docs/models) for descriptions of
+    them.
     """
 
     name: Optional[str] = None
@@ -85,16 +85,16 @@ class Assistant(BaseModel):
     response_format: Optional[AssistantResponseFormatOption] = None
     """Specifies the format that the model must output.
 
-    Compatible with [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-    [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+    Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+    [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
     and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
     Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-    Outputs which guarantees the model will match your supplied JSON schema. Learn
-    more in the
+    Outputs which ensures the model will match your supplied JSON schema. Learn more
+    in the
     [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-    Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+    Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
     message the model generates is valid JSON.
 
     **Important:** when using JSON mode, you **must** also instruct the model to
diff --git a/portkey_ai/_vendor/openai/types/beta/assistant_create_params.py b/portkey_ai/_vendor/openai/types/beta/assistant_create_params.py
index 84cd4425..568b223c 100644
--- a/portkey_ai/_vendor/openai/types/beta/assistant_create_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/assistant_create_params.py
@@ -3,10 +3,11 @@
 from __future__ import annotations
 
 from typing import List, Union, Iterable, Optional
-from typing_extensions import Literal, Required, TypeAlias, TypedDict
+from typing_extensions import Required, TypedDict
 
 from ..chat_model import ChatModel
 from .assistant_tool_param import AssistantToolParam
+from .file_chunking_strategy_param import FileChunkingStrategyParam
 from .assistant_response_format_option_param import AssistantResponseFormatOptionParam
 
 __all__ = [
@@ -15,10 +16,6 @@
     "ToolResourcesCodeInterpreter",
     "ToolResourcesFileSearch",
     "ToolResourcesFileSearchVectorStore",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategy",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategyAuto",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategyStatic",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic",
 ]
 
 
@@ -29,8 +26,8 @@ class AssistantCreateParams(TypedDict, total=False):
     You can use the
     [List models](https://platform.openai.com/docs/api-reference/models/list) API to
     see all of your available models, or see our
-    [Model overview](https://platform.openai.com/docs/models/overview) for
-    descriptions of them.
+    [Model overview](https://platform.openai.com/docs/models) for descriptions of
+    them.
     """
 
     description: Optional[str]
@@ -47,7 +44,7 @@ class AssistantCreateParams(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     name: Optional[str]
@@ -56,16 +53,16 @@ class AssistantCreateParams(TypedDict, total=False):
     response_format: Optional[AssistantResponseFormatOptionParam]
     """Specifies the format that the model must output.
 
-    Compatible with [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-    [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+    Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+    [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
     and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
     Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-    Outputs which guarantees the model will match your supplied JSON schema. Learn
-    more in the
+    Outputs which ensures the model will match your supplied JSON schema. Learn more
+    in the
     [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-    Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+    Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
     message the model generates is valid JSON.
 
     **Important:** when using JSON mode, you **must** also instruct the model to
@@ -118,43 +115,12 @@ class ToolResourcesCodeInterpreter(TypedDict, total=False):
     """
 
 
-class ToolResourcesFileSearchVectorStoreChunkingStrategyAuto(TypedDict, total=False):
-    type: Required[Literal["auto"]]
-    """Always `auto`."""
-
-
-class ToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic(TypedDict, total=False):
-    chunk_overlap_tokens: Required[int]
-    """The number of tokens that overlap between chunks. The default value is `400`.
-
-    Note that the overlap must not exceed half of `max_chunk_size_tokens`.
-    """
-
-    max_chunk_size_tokens: Required[int]
-    """The maximum number of tokens in each chunk.
-
-    The default value is `800`. The minimum value is `100` and the maximum value is
-    `4096`.
-    """
-
-
-class ToolResourcesFileSearchVectorStoreChunkingStrategyStatic(TypedDict, total=False):
-    static: Required[ToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic]
-
-    type: Required[Literal["static"]]
-    """Always `static`."""
-
-
-ToolResourcesFileSearchVectorStoreChunkingStrategy: TypeAlias = Union[
-    ToolResourcesFileSearchVectorStoreChunkingStrategyAuto, ToolResourcesFileSearchVectorStoreChunkingStrategyStatic
-]
-
-
 class ToolResourcesFileSearchVectorStore(TypedDict, total=False):
-    chunking_strategy: ToolResourcesFileSearchVectorStoreChunkingStrategy
+    chunking_strategy: FileChunkingStrategyParam
     """The chunking strategy used to chunk the file(s).
 
-    If not set, will use the `auto` strategy.
+    If not set, will use the `auto` strategy. Only applicable if `file_ids` is
+    non-empty.
     """
 
     file_ids: List[str]
@@ -169,7 +135,7 @@ class ToolResourcesFileSearchVectorStore(TypedDict, total=False):
 
     This can be useful for storing additional information about the vector store in
     a structured format. Keys can be a maximum of 64 characters long and values can
-    be a maxium of 512 characters long.
+    be a maximum of 512 characters long.
     """
 
 
diff --git a/portkey_ai/_vendor/openai/types/beta/assistant_list_params.py b/portkey_ai/_vendor/openai/types/beta/assistant_list_params.py
index f54f6312..834ffbca 100644
--- a/portkey_ai/_vendor/openai/types/beta/assistant_list_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/assistant_list_params.py
@@ -21,7 +21,7 @@ class AssistantListParams(TypedDict, total=False):
     """A cursor for use in pagination.
 
     `before` is an object ID that defines your place in the list. For instance, if
-    you make a list request and receive 100 objects, ending with obj_foo, your
+    you make a list request and receive 100 objects, starting with obj_foo, your
     subsequent call can include before=obj_foo in order to fetch the previous page
     of the list.
     """
diff --git a/portkey_ai/_vendor/openai/types/beta/assistant_response_format_option_param.py b/portkey_ai/_vendor/openai/types/beta/assistant_response_format_option_param.py
index 680a060c..5e724a4d 100644
--- a/portkey_ai/_vendor/openai/types/beta/assistant_response_format_option_param.py
+++ b/portkey_ai/_vendor/openai/types/beta/assistant_response_format_option_param.py
@@ -5,13 +5,12 @@
 from typing import Union
 from typing_extensions import Literal, TypeAlias
 
-from ...types import shared_params
+from ..shared_params.response_format_text import ResponseFormatText
+from ..shared_params.response_format_json_object import ResponseFormatJSONObject
+from ..shared_params.response_format_json_schema import ResponseFormatJSONSchema
 
 __all__ = ["AssistantResponseFormatOptionParam"]
 
 AssistantResponseFormatOptionParam: TypeAlias = Union[
-    Literal["auto"],
-    shared_params.ResponseFormatText,
-    shared_params.ResponseFormatJSONObject,
-    shared_params.ResponseFormatJSONSchema,
+    Literal["auto"], ResponseFormatText, ResponseFormatJSONObject, ResponseFormatJSONSchema
 ]
diff --git a/portkey_ai/_vendor/openai/types/beta/assistant_stream_event.py b/portkey_ai/_vendor/openai/types/beta/assistant_stream_event.py
index f1d8898f..41d3a0c5 100644
--- a/portkey_ai/_vendor/openai/types/beta/assistant_stream_event.py
+++ b/portkey_ai/_vendor/openai/types/beta/assistant_stream_event.py
@@ -1,6 +1,6 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing import Union
+from typing import Union, Optional
 from typing_extensions import Literal, Annotated, TypeAlias
 
 from .thread import Thread
@@ -51,6 +51,9 @@ class ThreadCreated(BaseModel):
 
     event: Literal["thread.created"]
 
+    enabled: Optional[bool] = None
+    """Whether to enable input audio transcription."""
+
 
 class ThreadRunCreated(BaseModel):
     data: Run
diff --git a/portkey_ai/_vendor/openai/types/beta/assistant_update_params.py b/portkey_ai/_vendor/openai/types/beta/assistant_update_params.py
index ade56581..9a66e41a 100644
--- a/portkey_ai/_vendor/openai/types/beta/assistant_update_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/assistant_update_params.py
@@ -26,7 +26,7 @@ class AssistantUpdateParams(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     model: str
@@ -35,8 +35,8 @@ class AssistantUpdateParams(TypedDict, total=False):
     You can use the
     [List models](https://platform.openai.com/docs/api-reference/models/list) API to
     see all of your available models, or see our
-    [Model overview](https://platform.openai.com/docs/models/overview) for
-    descriptions of them.
+    [Model overview](https://platform.openai.com/docs/models) for descriptions of
+    them.
     """
 
     name: Optional[str]
@@ -45,16 +45,16 @@ class AssistantUpdateParams(TypedDict, total=False):
     response_format: Optional[AssistantResponseFormatOptionParam]
     """Specifies the format that the model must output.
 
-    Compatible with [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-    [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+    Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+    [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
     and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
     Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-    Outputs which guarantees the model will match your supplied JSON schema. Learn
-    more in the
+    Outputs which ensures the model will match your supplied JSON schema. Learn more
+    in the
     [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-    Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+    Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
     message the model generates is valid JSON.
 
     **Important:** when using JSON mode, you **must** also instruct the model to
diff --git a/portkey_ai/_vendor/openai/types/beta/auto_file_chunking_strategy_param.py b/portkey_ai/_vendor/openai/types/beta/auto_file_chunking_strategy_param.py
new file mode 100644
index 00000000..6f17836b
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/auto_file_chunking_strategy_param.py
@@ -0,0 +1,12 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["AutoFileChunkingStrategyParam"]
+
+
+class AutoFileChunkingStrategyParam(TypedDict, total=False):
+    type: Required[Literal["auto"]]
+    """Always `auto`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/file_chunking_strategy.py b/portkey_ai/_vendor/openai/types/beta/file_chunking_strategy.py
new file mode 100644
index 00000000..406d69dd
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/file_chunking_strategy.py
@@ -0,0 +1,14 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Union
+from typing_extensions import Annotated, TypeAlias
+
+from ..._utils import PropertyInfo
+from .other_file_chunking_strategy_object import OtherFileChunkingStrategyObject
+from .static_file_chunking_strategy_object import StaticFileChunkingStrategyObject
+
+__all__ = ["FileChunkingStrategy"]
+
+FileChunkingStrategy: TypeAlias = Annotated[
+    Union[StaticFileChunkingStrategyObject, OtherFileChunkingStrategyObject], PropertyInfo(discriminator="type")
+]
diff --git a/portkey_ai/_vendor/openai/types/beta/file_chunking_strategy_param.py b/portkey_ai/_vendor/openai/types/beta/file_chunking_strategy_param.py
new file mode 100644
index 00000000..46383358
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/file_chunking_strategy_param.py
@@ -0,0 +1,13 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union
+from typing_extensions import TypeAlias
+
+from .auto_file_chunking_strategy_param import AutoFileChunkingStrategyParam
+from .static_file_chunking_strategy_param import StaticFileChunkingStrategyParam
+
+__all__ = ["FileChunkingStrategyParam"]
+
+FileChunkingStrategyParam: TypeAlias = Union[AutoFileChunkingStrategyParam, StaticFileChunkingStrategyParam]
diff --git a/portkey_ai/_vendor/openai/types/beta/file_search_tool.py b/portkey_ai/_vendor/openai/types/beta/file_search_tool.py
index 26ab1cb8..89fc16c0 100644
--- a/portkey_ai/_vendor/openai/types/beta/file_search_tool.py
+++ b/portkey_ai/_vendor/openai/types/beta/file_search_tool.py
@@ -5,7 +5,21 @@
 
 from ..._models import BaseModel
 
-__all__ = ["FileSearchTool", "FileSearch"]
+__all__ = ["FileSearchTool", "FileSearch", "FileSearchRankingOptions"]
+
+
+class FileSearchRankingOptions(BaseModel):
+    score_threshold: float
+    """The score threshold for the file search.
+
+    All values must be a floating point number between 0 and 1.
+    """
+
+    ranker: Optional[Literal["auto", "default_2024_08_21"]] = None
+    """The ranker to use for the file search.
+
+    If not specified will use the `auto` ranker.
+    """
 
 
 class FileSearch(BaseModel):
@@ -17,7 +31,18 @@ class FileSearch(BaseModel):
 
     Note that the file search tool may output fewer than `max_num_results` results.
     See the
-    [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search/number-of-chunks-returned)
+    [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+    for more information.
+    """
+
+    ranking_options: Optional[FileSearchRankingOptions] = None
+    """The ranking options for the file search.
+
+    If not specified, the file search tool will use the `auto` ranker and a
+    score_threshold of 0.
+
+    See the
+    [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
     for more information.
     """
 
diff --git a/portkey_ai/_vendor/openai/types/beta/file_search_tool_param.py b/portkey_ai/_vendor/openai/types/beta/file_search_tool_param.py
index 666719f8..c73d0af7 100644
--- a/portkey_ai/_vendor/openai/types/beta/file_search_tool_param.py
+++ b/portkey_ai/_vendor/openai/types/beta/file_search_tool_param.py
@@ -4,7 +4,21 @@
 
 from typing_extensions import Literal, Required, TypedDict
 
-__all__ = ["FileSearchToolParam", "FileSearch"]
+__all__ = ["FileSearchToolParam", "FileSearch", "FileSearchRankingOptions"]
+
+
+class FileSearchRankingOptions(TypedDict, total=False):
+    score_threshold: Required[float]
+    """The score threshold for the file search.
+
+    All values must be a floating point number between 0 and 1.
+    """
+
+    ranker: Literal["auto", "default_2024_08_21"]
+    """The ranker to use for the file search.
+
+    If not specified will use the `auto` ranker.
+    """
 
 
 class FileSearch(TypedDict, total=False):
@@ -16,7 +30,18 @@ class FileSearch(TypedDict, total=False):
 
     Note that the file search tool may output fewer than `max_num_results` results.
     See the
-    [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search/number-of-chunks-returned)
+    [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+    for more information.
+    """
+
+    ranking_options: FileSearchRankingOptions
+    """The ranking options for the file search.
+
+    If not specified, the file search tool will use the `auto` ranker and a
+    score_threshold of 0.
+
+    See the
+    [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
     for more information.
     """
 
diff --git a/portkey_ai/_vendor/openai/types/beta/function_tool_param.py b/portkey_ai/_vendor/openai/types/beta/function_tool_param.py
index b44c0d47..d906e02b 100644
--- a/portkey_ai/_vendor/openai/types/beta/function_tool_param.py
+++ b/portkey_ai/_vendor/openai/types/beta/function_tool_param.py
@@ -4,13 +4,13 @@
 
 from typing_extensions import Literal, Required, TypedDict
 
-from ...types import shared_params
+from ..shared_params.function_definition import FunctionDefinition
 
 __all__ = ["FunctionToolParam"]
 
 
 class FunctionToolParam(TypedDict, total=False):
-    function: Required[shared_params.FunctionDefinition]
+    function: Required[FunctionDefinition]
 
     type: Required[Literal["function"]]
     """The type of tool being defined: `function`"""
diff --git a/portkey_ai/_vendor/openai/types/beta/other_file_chunking_strategy_object.py b/portkey_ai/_vendor/openai/types/beta/other_file_chunking_strategy_object.py
new file mode 100644
index 00000000..89da560b
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/other_file_chunking_strategy_object.py
@@ -0,0 +1,12 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ..._models import BaseModel
+
+__all__ = ["OtherFileChunkingStrategyObject"]
+
+
+class OtherFileChunkingStrategyObject(BaseModel):
+    type: Literal["other"]
+    """Always `other`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/__init__.py b/portkey_ai/_vendor/openai/types/beta/realtime/__init__.py
new file mode 100644
index 00000000..372d4ec1
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/__init__.py
@@ -0,0 +1,80 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from .session import Session as Session
+from .error_event import ErrorEvent as ErrorEvent
+from .conversation_item import ConversationItem as ConversationItem
+from .realtime_response import RealtimeResponse as RealtimeResponse
+from .response_done_event import ResponseDoneEvent as ResponseDoneEvent
+from .session_update_event import SessionUpdateEvent as SessionUpdateEvent
+from .realtime_client_event import RealtimeClientEvent as RealtimeClientEvent
+from .realtime_server_event import RealtimeServerEvent as RealtimeServerEvent
+from .response_cancel_event import ResponseCancelEvent as ResponseCancelEvent
+from .response_create_event import ResponseCreateEvent as ResponseCreateEvent
+from .session_create_params import SessionCreateParams as SessionCreateParams
+from .session_created_event import SessionCreatedEvent as SessionCreatedEvent
+from .session_updated_event import SessionUpdatedEvent as SessionUpdatedEvent
+from .response_created_event import ResponseCreatedEvent as ResponseCreatedEvent
+from .conversation_item_param import ConversationItemParam as ConversationItemParam
+from .realtime_connect_params import RealtimeConnectParams as RealtimeConnectParams
+from .realtime_response_usage import RealtimeResponseUsage as RealtimeResponseUsage
+from .session_create_response import SessionCreateResponse as SessionCreateResponse
+from .realtime_response_status import RealtimeResponseStatus as RealtimeResponseStatus
+from .response_text_done_event import ResponseTextDoneEvent as ResponseTextDoneEvent
+from .conversation_item_content import ConversationItemContent as ConversationItemContent
+from .rate_limits_updated_event import RateLimitsUpdatedEvent as RateLimitsUpdatedEvent
+from .response_audio_done_event import ResponseAudioDoneEvent as ResponseAudioDoneEvent
+from .response_text_delta_event import ResponseTextDeltaEvent as ResponseTextDeltaEvent
+from .conversation_created_event import ConversationCreatedEvent as ConversationCreatedEvent
+from .response_audio_delta_event import ResponseAudioDeltaEvent as ResponseAudioDeltaEvent
+from .session_update_event_param import SessionUpdateEventParam as SessionUpdateEventParam
+from .realtime_client_event_param import RealtimeClientEventParam as RealtimeClientEventParam
+from .response_cancel_event_param import ResponseCancelEventParam as ResponseCancelEventParam
+from .response_create_event_param import ResponseCreateEventParam as ResponseCreateEventParam
+from .conversation_item_create_event import ConversationItemCreateEvent as ConversationItemCreateEvent
+from .conversation_item_delete_event import ConversationItemDeleteEvent as ConversationItemDeleteEvent
+from .input_audio_buffer_clear_event import InputAudioBufferClearEvent as InputAudioBufferClearEvent
+from .conversation_item_content_param import ConversationItemContentParam as ConversationItemContentParam
+from .conversation_item_created_event import ConversationItemCreatedEvent as ConversationItemCreatedEvent
+from .conversation_item_deleted_event import ConversationItemDeletedEvent as ConversationItemDeletedEvent
+from .input_audio_buffer_append_event import InputAudioBufferAppendEvent as InputAudioBufferAppendEvent
+from .input_audio_buffer_commit_event import InputAudioBufferCommitEvent as InputAudioBufferCommitEvent
+from .response_output_item_done_event import ResponseOutputItemDoneEvent as ResponseOutputItemDoneEvent
+from .conversation_item_truncate_event import ConversationItemTruncateEvent as ConversationItemTruncateEvent
+from .input_audio_buffer_cleared_event import InputAudioBufferClearedEvent as InputAudioBufferClearedEvent
+from .response_content_part_done_event import ResponseContentPartDoneEvent as ResponseContentPartDoneEvent
+from .response_output_item_added_event import ResponseOutputItemAddedEvent as ResponseOutputItemAddedEvent
+from .conversation_item_truncated_event import ConversationItemTruncatedEvent as ConversationItemTruncatedEvent
+from .response_content_part_added_event import ResponseContentPartAddedEvent as ResponseContentPartAddedEvent
+from .input_audio_buffer_committed_event import InputAudioBufferCommittedEvent as InputAudioBufferCommittedEvent
+from .conversation_item_create_event_param import ConversationItemCreateEventParam as ConversationItemCreateEventParam
+from .conversation_item_delete_event_param import ConversationItemDeleteEventParam as ConversationItemDeleteEventParam
+from .input_audio_buffer_clear_event_param import InputAudioBufferClearEventParam as InputAudioBufferClearEventParam
+from .response_audio_transcript_done_event import ResponseAudioTranscriptDoneEvent as ResponseAudioTranscriptDoneEvent
+from .input_audio_buffer_append_event_param import InputAudioBufferAppendEventParam as InputAudioBufferAppendEventParam
+from .input_audio_buffer_commit_event_param import InputAudioBufferCommitEventParam as InputAudioBufferCommitEventParam
+from .response_audio_transcript_delta_event import (
+    ResponseAudioTranscriptDeltaEvent as ResponseAudioTranscriptDeltaEvent,
+)
+from .conversation_item_truncate_event_param import (
+    ConversationItemTruncateEventParam as ConversationItemTruncateEventParam,
+)
+from .input_audio_buffer_speech_started_event import (
+    InputAudioBufferSpeechStartedEvent as InputAudioBufferSpeechStartedEvent,
+)
+from .input_audio_buffer_speech_stopped_event import (
+    InputAudioBufferSpeechStoppedEvent as InputAudioBufferSpeechStoppedEvent,
+)
+from .response_function_call_arguments_done_event import (
+    ResponseFunctionCallArgumentsDoneEvent as ResponseFunctionCallArgumentsDoneEvent,
+)
+from .response_function_call_arguments_delta_event import (
+    ResponseFunctionCallArgumentsDeltaEvent as ResponseFunctionCallArgumentsDeltaEvent,
+)
+from .conversation_item_input_audio_transcription_failed_event import (
+    ConversationItemInputAudioTranscriptionFailedEvent as ConversationItemInputAudioTranscriptionFailedEvent,
+)
+from .conversation_item_input_audio_transcription_completed_event import (
+    ConversationItemInputAudioTranscriptionCompletedEvent as ConversationItemInputAudioTranscriptionCompletedEvent,
+)
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_created_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_created_event.py
new file mode 100644
index 00000000..4ba05408
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_created_event.py
@@ -0,0 +1,27 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ConversationCreatedEvent", "Conversation"]
+
+
+class Conversation(BaseModel):
+    id: Optional[str] = None
+    """The unique ID of the conversation."""
+
+    object: Optional[Literal["realtime.conversation"]] = None
+    """The object type, must be `realtime.conversation`."""
+
+
+class ConversationCreatedEvent(BaseModel):
+    conversation: Conversation
+    """The conversation resource."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    type: Literal["conversation.created"]
+    """The event type, must be `conversation.created`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item.py
new file mode 100644
index 00000000..4edf6c4d
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item.py
@@ -0,0 +1,61 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List, Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+from .conversation_item_content import ConversationItemContent
+
+__all__ = ["ConversationItem"]
+
+
+class ConversationItem(BaseModel):
+    id: Optional[str] = None
+    """
+    The unique ID of the item, this can be generated by the client to help manage
+    server-side context, but is not required because the server will generate one if
+    not provided.
+    """
+
+    arguments: Optional[str] = None
+    """The arguments of the function call (for `function_call` items)."""
+
+    call_id: Optional[str] = None
+    """
+    The ID of the function call (for `function_call` and `function_call_output`
+    items). If passed on a `function_call_output` item, the server will check that a
+    `function_call` item with the same ID exists in the conversation history.
+    """
+
+    content: Optional[List[ConversationItemContent]] = None
+    """The content of the message, applicable for `message` items.
+
+    - Message items of role `system` support only `input_text` content
+    - Message items of role `user` support `input_text` and `input_audio` content
+    - Message items of role `assistant` support `text` content.
+    """
+
+    name: Optional[str] = None
+    """The name of the function being called (for `function_call` items)."""
+
+    object: Optional[Literal["realtime.item"]] = None
+    """Identifier for the API object being returned - always `realtime.item`."""
+
+    output: Optional[str] = None
+    """The output of the function call (for `function_call_output` items)."""
+
+    role: Optional[Literal["user", "assistant", "system"]] = None
+    """
+    The role of the message sender (`user`, `assistant`, `system`), only applicable
+    for `message` items.
+    """
+
+    status: Optional[Literal["completed", "incomplete"]] = None
+    """The status of the item (`completed`, `incomplete`).
+
+    These have no effect on the conversation, but are accepted for consistency with
+    the `conversation.item.created` event.
+    """
+
+    type: Optional[Literal["message", "function_call", "function_call_output"]] = None
+    """The type of the item (`message`, `function_call`, `function_call_output`)."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_content.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_content.py
new file mode 100644
index 00000000..b854aa0e
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_content.py
@@ -0,0 +1,28 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ConversationItemContent"]
+
+
+class ConversationItemContent(BaseModel):
+    id: Optional[str] = None
+    """
+    ID of a previous conversation item (like a model response), used for
+    `item_reference` content types.
+    """
+
+    audio: Optional[str] = None
+    """Base64-encoded audio bytes, used for `input_audio` content type."""
+
+    text: Optional[str] = None
+    """The text content, used for `input_text` and `text` content types."""
+
+    transcript: Optional[str] = None
+    """The transcript of the audio, used for `input_audio` content type."""
+
+    type: Optional[Literal["input_text", "input_audio", "item_reference", "text"]] = None
+    """The content type (`input_text`, `input_audio`, `item_reference`, `text`)."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_content_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_content_param.py
new file mode 100644
index 00000000..b354d789
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_content_param.py
@@ -0,0 +1,27 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, TypedDict
+
+__all__ = ["ConversationItemContentParam"]
+
+
+class ConversationItemContentParam(TypedDict, total=False):
+    id: str
+    """
+    ID of a previous conversation item (like a model response), used for
+    `item_reference` content types.
+    """
+
+    audio: str
+    """Base64-encoded audio bytes, used for `input_audio` content type."""
+
+    text: str
+    """The text content, used for `input_text` and `text` content types."""
+
+    transcript: str
+    """The transcript of the audio, used for `input_audio` content type."""
+
+    type: Literal["input_text", "input_audio", "item_reference", "text"]
+    """The content type (`input_text`, `input_audio`, `item_reference`, `text`)."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_create_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_create_event.py
new file mode 100644
index 00000000..50d30967
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_create_event.py
@@ -0,0 +1,28 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+from .conversation_item import ConversationItem
+
+__all__ = ["ConversationItemCreateEvent"]
+
+
+class ConversationItemCreateEvent(BaseModel):
+    item: ConversationItem
+    """The item to add to the conversation."""
+
+    type: Literal["conversation.item.create"]
+    """The event type, must be `conversation.item.create`."""
+
+    event_id: Optional[str] = None
+    """Optional client-generated ID used to identify this event."""
+
+    previous_item_id: Optional[str] = None
+    """The ID of the preceding item after which the new item will be inserted.
+
+    If not set, the new item will be appended to the end of the conversation. If
+    set, it allows an item to be inserted mid-conversation. If the ID cannot be
+    found, an error will be returned and the item will not be added.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_create_event_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_create_event_param.py
new file mode 100644
index 00000000..b8c8bbc2
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_create_event_param.py
@@ -0,0 +1,28 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+from .conversation_item_param import ConversationItemParam
+
+__all__ = ["ConversationItemCreateEventParam"]
+
+
+class ConversationItemCreateEventParam(TypedDict, total=False):
+    item: Required[ConversationItemParam]
+    """The item to add to the conversation."""
+
+    type: Required[Literal["conversation.item.create"]]
+    """The event type, must be `conversation.item.create`."""
+
+    event_id: str
+    """Optional client-generated ID used to identify this event."""
+
+    previous_item_id: str
+    """The ID of the preceding item after which the new item will be inserted.
+
+    If not set, the new item will be appended to the end of the conversation. If
+    set, it allows an item to be inserted mid-conversation. If the ID cannot be
+    found, an error will be returned and the item will not be added.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_created_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_created_event.py
new file mode 100644
index 00000000..2f203882
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_created_event.py
@@ -0,0 +1,25 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+from .conversation_item import ConversationItem
+
+__all__ = ["ConversationItemCreatedEvent"]
+
+
+class ConversationItemCreatedEvent(BaseModel):
+    event_id: str
+    """The unique ID of the server event."""
+
+    item: ConversationItem
+    """The item to add to the conversation."""
+
+    previous_item_id: str
+    """
+    The ID of the preceding item in the Conversation context, allows the client to
+    understand the order of the conversation.
+    """
+
+    type: Literal["conversation.item.created"]
+    """The event type, must be `conversation.item.created`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_delete_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_delete_event.py
new file mode 100644
index 00000000..02ca8250
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_delete_event.py
@@ -0,0 +1,19 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ConversationItemDeleteEvent"]
+
+
+class ConversationItemDeleteEvent(BaseModel):
+    item_id: str
+    """The ID of the item to delete."""
+
+    type: Literal["conversation.item.delete"]
+    """The event type, must be `conversation.item.delete`."""
+
+    event_id: Optional[str] = None
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_delete_event_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_delete_event_param.py
new file mode 100644
index 00000000..c3f88d66
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_delete_event_param.py
@@ -0,0 +1,18 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["ConversationItemDeleteEventParam"]
+
+
+class ConversationItemDeleteEventParam(TypedDict, total=False):
+    item_id: Required[str]
+    """The ID of the item to delete."""
+
+    type: Required[Literal["conversation.item.delete"]]
+    """The event type, must be `conversation.item.delete`."""
+
+    event_id: str
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_deleted_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_deleted_event.py
new file mode 100644
index 00000000..a35a9781
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_deleted_event.py
@@ -0,0 +1,18 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ConversationItemDeletedEvent"]
+
+
+class ConversationItemDeletedEvent(BaseModel):
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the item that was deleted."""
+
+    type: Literal["conversation.item.deleted"]
+    """The event type, must be `conversation.item.deleted`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_input_audio_transcription_completed_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_input_audio_transcription_completed_event.py
new file mode 100644
index 00000000..ded79cc0
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_input_audio_transcription_completed_event.py
@@ -0,0 +1,26 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ConversationItemInputAudioTranscriptionCompletedEvent"]
+
+
+class ConversationItemInputAudioTranscriptionCompletedEvent(BaseModel):
+    content_index: int
+    """The index of the content part containing the audio."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the user message item containing the audio."""
+
+    transcript: str
+    """The transcribed text."""
+
+    type: Literal["conversation.item.input_audio_transcription.completed"]
+    """
+    The event type, must be `conversation.item.input_audio_transcription.completed`.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_input_audio_transcription_failed_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_input_audio_transcription_failed_event.py
new file mode 100644
index 00000000..cecac93e
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_input_audio_transcription_failed_event.py
@@ -0,0 +1,39 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ConversationItemInputAudioTranscriptionFailedEvent", "Error"]
+
+
+class Error(BaseModel):
+    code: Optional[str] = None
+    """Error code, if any."""
+
+    message: Optional[str] = None
+    """A human-readable error message."""
+
+    param: Optional[str] = None
+    """Parameter related to the error, if any."""
+
+    type: Optional[str] = None
+    """The type of error."""
+
+
+class ConversationItemInputAudioTranscriptionFailedEvent(BaseModel):
+    content_index: int
+    """The index of the content part containing the audio."""
+
+    error: Error
+    """Details of the transcription error."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the user message item."""
+
+    type: Literal["conversation.item.input_audio_transcription.failed"]
+    """The event type, must be `conversation.item.input_audio_transcription.failed`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_param.py
new file mode 100644
index 00000000..ac0f8431
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_param.py
@@ -0,0 +1,62 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Iterable
+from typing_extensions import Literal, TypedDict
+
+from .conversation_item_content_param import ConversationItemContentParam
+
+__all__ = ["ConversationItemParam"]
+
+
+class ConversationItemParam(TypedDict, total=False):
+    id: str
+    """
+    The unique ID of the item, this can be generated by the client to help manage
+    server-side context, but is not required because the server will generate one if
+    not provided.
+    """
+
+    arguments: str
+    """The arguments of the function call (for `function_call` items)."""
+
+    call_id: str
+    """
+    The ID of the function call (for `function_call` and `function_call_output`
+    items). If passed on a `function_call_output` item, the server will check that a
+    `function_call` item with the same ID exists in the conversation history.
+    """
+
+    content: Iterable[ConversationItemContentParam]
+    """The content of the message, applicable for `message` items.
+
+    - Message items of role `system` support only `input_text` content
+    - Message items of role `user` support `input_text` and `input_audio` content
+    - Message items of role `assistant` support `text` content.
+    """
+
+    name: str
+    """The name of the function being called (for `function_call` items)."""
+
+    object: Literal["realtime.item"]
+    """Identifier for the API object being returned - always `realtime.item`."""
+
+    output: str
+    """The output of the function call (for `function_call_output` items)."""
+
+    role: Literal["user", "assistant", "system"]
+    """
+    The role of the message sender (`user`, `assistant`, `system`), only applicable
+    for `message` items.
+    """
+
+    status: Literal["completed", "incomplete"]
+    """The status of the item (`completed`, `incomplete`).
+
+    These have no effect on the conversation, but are accepted for consistency with
+    the `conversation.item.created` event.
+    """
+
+    type: Literal["message", "function_call", "function_call_output"]
+    """The type of the item (`message`, `function_call`, `function_call_output`)."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_truncate_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_truncate_event.py
new file mode 100644
index 00000000..cb336bba
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_truncate_event.py
@@ -0,0 +1,32 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ConversationItemTruncateEvent"]
+
+
+class ConversationItemTruncateEvent(BaseModel):
+    audio_end_ms: int
+    """Inclusive duration up to which audio is truncated, in milliseconds.
+
+    If the audio_end_ms is greater than the actual audio duration, the server will
+    respond with an error.
+    """
+
+    content_index: int
+    """The index of the content part to truncate. Set this to 0."""
+
+    item_id: str
+    """The ID of the assistant message item to truncate.
+
+    Only assistant message items can be truncated.
+    """
+
+    type: Literal["conversation.item.truncate"]
+    """The event type, must be `conversation.item.truncate`."""
+
+    event_id: Optional[str] = None
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_truncate_event_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_truncate_event_param.py
new file mode 100644
index 00000000..d3ad1e1e
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_truncate_event_param.py
@@ -0,0 +1,31 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["ConversationItemTruncateEventParam"]
+
+
+class ConversationItemTruncateEventParam(TypedDict, total=False):
+    audio_end_ms: Required[int]
+    """Inclusive duration up to which audio is truncated, in milliseconds.
+
+    If the audio_end_ms is greater than the actual audio duration, the server will
+    respond with an error.
+    """
+
+    content_index: Required[int]
+    """The index of the content part to truncate. Set this to 0."""
+
+    item_id: Required[str]
+    """The ID of the assistant message item to truncate.
+
+    Only assistant message items can be truncated.
+    """
+
+    type: Required[Literal["conversation.item.truncate"]]
+    """The event type, must be `conversation.item.truncate`."""
+
+    event_id: str
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_truncated_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_truncated_event.py
new file mode 100644
index 00000000..36368fa2
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/conversation_item_truncated_event.py
@@ -0,0 +1,24 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ConversationItemTruncatedEvent"]
+
+
+class ConversationItemTruncatedEvent(BaseModel):
+    audio_end_ms: int
+    """The duration up to which the audio was truncated, in milliseconds."""
+
+    content_index: int
+    """The index of the content part that was truncated."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the assistant message item that was truncated."""
+
+    type: Literal["conversation.item.truncated"]
+    """The event type, must be `conversation.item.truncated`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/error_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/error_event.py
new file mode 100644
index 00000000..e020fc38
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/error_event.py
@@ -0,0 +1,36 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ErrorEvent", "Error"]
+
+
+class Error(BaseModel):
+    message: str
+    """A human-readable error message."""
+
+    type: str
+    """The type of error (e.g., "invalid_request_error", "server_error")."""
+
+    code: Optional[str] = None
+    """Error code, if any."""
+
+    event_id: Optional[str] = None
+    """The event_id of the client event that caused the error, if applicable."""
+
+    param: Optional[str] = None
+    """Parameter related to the error, if any."""
+
+
+class ErrorEvent(BaseModel):
+    error: Error
+    """Details of the error."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    type: Literal["error"]
+    """The event type, must be `error`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_append_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_append_event.py
new file mode 100644
index 00000000..a253a648
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_append_event.py
@@ -0,0 +1,23 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["InputAudioBufferAppendEvent"]
+
+
+class InputAudioBufferAppendEvent(BaseModel):
+    audio: str
+    """Base64-encoded audio bytes.
+
+    This must be in the format specified by the `input_audio_format` field in the
+    session configuration.
+    """
+
+    type: Literal["input_audio_buffer.append"]
+    """The event type, must be `input_audio_buffer.append`."""
+
+    event_id: Optional[str] = None
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_append_event_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_append_event_param.py
new file mode 100644
index 00000000..3ad0bc73
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_append_event_param.py
@@ -0,0 +1,22 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["InputAudioBufferAppendEventParam"]
+
+
+class InputAudioBufferAppendEventParam(TypedDict, total=False):
+    audio: Required[str]
+    """Base64-encoded audio bytes.
+
+    This must be in the format specified by the `input_audio_format` field in the
+    session configuration.
+    """
+
+    type: Required[Literal["input_audio_buffer.append"]]
+    """The event type, must be `input_audio_buffer.append`."""
+
+    event_id: str
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_clear_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_clear_event.py
new file mode 100644
index 00000000..b0624d34
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_clear_event.py
@@ -0,0 +1,16 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["InputAudioBufferClearEvent"]
+
+
+class InputAudioBufferClearEvent(BaseModel):
+    type: Literal["input_audio_buffer.clear"]
+    """The event type, must be `input_audio_buffer.clear`."""
+
+    event_id: Optional[str] = None
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_clear_event_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_clear_event_param.py
new file mode 100644
index 00000000..2bd6bc5a
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_clear_event_param.py
@@ -0,0 +1,15 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["InputAudioBufferClearEventParam"]
+
+
+class InputAudioBufferClearEventParam(TypedDict, total=False):
+    type: Required[Literal["input_audio_buffer.clear"]]
+    """The event type, must be `input_audio_buffer.clear`."""
+
+    event_id: str
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_cleared_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_cleared_event.py
new file mode 100644
index 00000000..632e1b94
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_cleared_event.py
@@ -0,0 +1,15 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["InputAudioBufferClearedEvent"]
+
+
+class InputAudioBufferClearedEvent(BaseModel):
+    event_id: str
+    """The unique ID of the server event."""
+
+    type: Literal["input_audio_buffer.cleared"]
+    """The event type, must be `input_audio_buffer.cleared`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_commit_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_commit_event.py
new file mode 100644
index 00000000..7b6f5e46
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_commit_event.py
@@ -0,0 +1,16 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["InputAudioBufferCommitEvent"]
+
+
+class InputAudioBufferCommitEvent(BaseModel):
+    type: Literal["input_audio_buffer.commit"]
+    """The event type, must be `input_audio_buffer.commit`."""
+
+    event_id: Optional[str] = None
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_commit_event_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_commit_event_param.py
new file mode 100644
index 00000000..c9c927ab
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_commit_event_param.py
@@ -0,0 +1,15 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["InputAudioBufferCommitEventParam"]
+
+
+class InputAudioBufferCommitEventParam(TypedDict, total=False):
+    type: Required[Literal["input_audio_buffer.commit"]]
+    """The event type, must be `input_audio_buffer.commit`."""
+
+    event_id: str
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_committed_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_committed_event.py
new file mode 100644
index 00000000..3071eff3
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_committed_event.py
@@ -0,0 +1,21 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["InputAudioBufferCommittedEvent"]
+
+
+class InputAudioBufferCommittedEvent(BaseModel):
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the user message item that will be created."""
+
+    previous_item_id: str
+    """The ID of the preceding item after which the new item will be inserted."""
+
+    type: Literal["input_audio_buffer.committed"]
+    """The event type, must be `input_audio_buffer.committed`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_speech_started_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_speech_started_event.py
new file mode 100644
index 00000000..4f3ab082
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_speech_started_event.py
@@ -0,0 +1,26 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["InputAudioBufferSpeechStartedEvent"]
+
+
+class InputAudioBufferSpeechStartedEvent(BaseModel):
+    audio_start_ms: int
+    """
+    Milliseconds from the start of all audio written to the buffer during the
+    session when speech was first detected. This will correspond to the beginning of
+    audio sent to the model, and thus includes the `prefix_padding_ms` configured in
+    the Session.
+    """
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the user message item that will be created when speech stops."""
+
+    type: Literal["input_audio_buffer.speech_started"]
+    """The event type, must be `input_audio_buffer.speech_started`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_speech_stopped_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_speech_stopped_event.py
new file mode 100644
index 00000000..40568170
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/input_audio_buffer_speech_stopped_event.py
@@ -0,0 +1,25 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["InputAudioBufferSpeechStoppedEvent"]
+
+
+class InputAudioBufferSpeechStoppedEvent(BaseModel):
+    audio_end_ms: int
+    """Milliseconds since the session started when speech stopped.
+
+    This will correspond to the end of audio sent to the model, and thus includes
+    the `min_silence_duration_ms` configured in the Session.
+    """
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the user message item that will be created."""
+
+    type: Literal["input_audio_buffer.speech_stopped"]
+    """The event type, must be `input_audio_buffer.speech_stopped`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/rate_limits_updated_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/rate_limits_updated_event.py
new file mode 100644
index 00000000..7e12283c
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/rate_limits_updated_event.py
@@ -0,0 +1,33 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List, Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["RateLimitsUpdatedEvent", "RateLimit"]
+
+
+class RateLimit(BaseModel):
+    limit: Optional[int] = None
+    """The maximum allowed value for the rate limit."""
+
+    name: Optional[Literal["requests", "tokens"]] = None
+    """The name of the rate limit (`requests`, `tokens`)."""
+
+    remaining: Optional[int] = None
+    """The remaining value before the limit is reached."""
+
+    reset_seconds: Optional[float] = None
+    """Seconds until the rate limit resets."""
+
+
+class RateLimitsUpdatedEvent(BaseModel):
+    event_id: str
+    """The unique ID of the server event."""
+
+    rate_limits: List[RateLimit]
+    """List of rate limit information."""
+
+    type: Literal["rate_limits.updated"]
+    """The event type, must be `rate_limits.updated`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/realtime_client_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_client_event.py
new file mode 100644
index 00000000..0769184c
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_client_event.py
@@ -0,0 +1,32 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Union
+from typing_extensions import Annotated, TypeAlias
+
+from ...._utils import PropertyInfo
+from .session_update_event import SessionUpdateEvent
+from .response_cancel_event import ResponseCancelEvent
+from .response_create_event import ResponseCreateEvent
+from .conversation_item_create_event import ConversationItemCreateEvent
+from .conversation_item_delete_event import ConversationItemDeleteEvent
+from .input_audio_buffer_clear_event import InputAudioBufferClearEvent
+from .input_audio_buffer_append_event import InputAudioBufferAppendEvent
+from .input_audio_buffer_commit_event import InputAudioBufferCommitEvent
+from .conversation_item_truncate_event import ConversationItemTruncateEvent
+
+__all__ = ["RealtimeClientEvent"]
+
+RealtimeClientEvent: TypeAlias = Annotated[
+    Union[
+        SessionUpdateEvent,
+        InputAudioBufferAppendEvent,
+        InputAudioBufferCommitEvent,
+        InputAudioBufferClearEvent,
+        ConversationItemCreateEvent,
+        ConversationItemTruncateEvent,
+        ConversationItemDeleteEvent,
+        ResponseCreateEvent,
+        ResponseCancelEvent,
+    ],
+    PropertyInfo(discriminator="type"),
+]
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/realtime_client_event_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_client_event_param.py
new file mode 100644
index 00000000..4020892c
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_client_event_param.py
@@ -0,0 +1,30 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union
+from typing_extensions import TypeAlias
+
+from .session_update_event_param import SessionUpdateEventParam
+from .response_cancel_event_param import ResponseCancelEventParam
+from .response_create_event_param import ResponseCreateEventParam
+from .conversation_item_create_event_param import ConversationItemCreateEventParam
+from .conversation_item_delete_event_param import ConversationItemDeleteEventParam
+from .input_audio_buffer_clear_event_param import InputAudioBufferClearEventParam
+from .input_audio_buffer_append_event_param import InputAudioBufferAppendEventParam
+from .input_audio_buffer_commit_event_param import InputAudioBufferCommitEventParam
+from .conversation_item_truncate_event_param import ConversationItemTruncateEventParam
+
+__all__ = ["RealtimeClientEventParam"]
+
+RealtimeClientEventParam: TypeAlias = Union[
+    SessionUpdateEventParam,
+    InputAudioBufferAppendEventParam,
+    InputAudioBufferCommitEventParam,
+    InputAudioBufferClearEventParam,
+    ConversationItemCreateEventParam,
+    ConversationItemTruncateEventParam,
+    ConversationItemDeleteEventParam,
+    ResponseCreateEventParam,
+    ResponseCancelEventParam,
+]
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/realtime_connect_params.py b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_connect_params.py
new file mode 100644
index 00000000..76474f3d
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_connect_params.py
@@ -0,0 +1,11 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Required, TypedDict
+
+__all__ = ["RealtimeConnectParams"]
+
+
+class RealtimeConnectParams(TypedDict, total=False):
+    model: Required[str]
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/realtime_response.py b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_response.py
new file mode 100644
index 00000000..3e1b1406
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_response.py
@@ -0,0 +1,42 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List, Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+from .conversation_item import ConversationItem
+from .realtime_response_usage import RealtimeResponseUsage
+from .realtime_response_status import RealtimeResponseStatus
+
+__all__ = ["RealtimeResponse"]
+
+
+class RealtimeResponse(BaseModel):
+    id: Optional[str] = None
+    """The unique ID of the response."""
+
+    metadata: Optional[object] = None
+    """Developer-provided string key-value pairs associated with this response."""
+
+    object: Optional[Literal["realtime.response"]] = None
+    """The object type, must be `realtime.response`."""
+
+    output: Optional[List[ConversationItem]] = None
+    """The list of output items generated by the response."""
+
+    status: Optional[Literal["completed", "cancelled", "failed", "incomplete"]] = None
+    """
+    The final status of the response (`completed`, `cancelled`, `failed`, or
+    `incomplete`).
+    """
+
+    status_details: Optional[RealtimeResponseStatus] = None
+    """Additional details about the status."""
+
+    usage: Optional[RealtimeResponseUsage] = None
+    """Usage statistics for the Response, this will correspond to billing.
+
+    A Realtime API session will maintain a conversation context and append new Items
+    to the Conversation, thus output from previous turns (text and audio tokens)
+    will become the input for later turns.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/realtime_response_status.py b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_response_status.py
new file mode 100644
index 00000000..7189cd58
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_response_status.py
@@ -0,0 +1,39 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["RealtimeResponseStatus", "Error"]
+
+
+class Error(BaseModel):
+    code: Optional[str] = None
+    """Error code, if any."""
+
+    type: Optional[str] = None
+    """The type of error."""
+
+
+class RealtimeResponseStatus(BaseModel):
+    error: Optional[Error] = None
+    """
+    A description of the error that caused the response to fail, populated when the
+    `status` is `failed`.
+    """
+
+    reason: Optional[Literal["turn_detected", "client_cancelled", "max_output_tokens", "content_filter"]] = None
+    """The reason the Response did not complete.
+
+    For a `cancelled` Response, one of `turn_detected` (the server VAD detected a
+    new start of speech) or `client_cancelled` (the client sent a cancel event). For
+    an `incomplete` Response, one of `max_output_tokens` or `content_filter` (the
+    server-side safety filter activated and cut off the response).
+    """
+
+    type: Optional[Literal["completed", "cancelled", "incomplete", "failed"]] = None
+    """
+    The type of error that caused the response to fail, corresponding with the
+    `status` field (`completed`, `cancelled`, `incomplete`, `failed`).
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/realtime_response_usage.py b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_response_usage.py
new file mode 100644
index 00000000..7ca822e2
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_response_usage.py
@@ -0,0 +1,52 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+
+from ...._models import BaseModel
+
+__all__ = ["RealtimeResponseUsage", "InputTokenDetails", "OutputTokenDetails"]
+
+
+class InputTokenDetails(BaseModel):
+    audio_tokens: Optional[int] = None
+    """The number of audio tokens used in the Response."""
+
+    cached_tokens: Optional[int] = None
+    """The number of cached tokens used in the Response."""
+
+    text_tokens: Optional[int] = None
+    """The number of text tokens used in the Response."""
+
+
+class OutputTokenDetails(BaseModel):
+    audio_tokens: Optional[int] = None
+    """The number of audio tokens used in the Response."""
+
+    text_tokens: Optional[int] = None
+    """The number of text tokens used in the Response."""
+
+
+class RealtimeResponseUsage(BaseModel):
+    input_token_details: Optional[InputTokenDetails] = None
+    """Details about the input tokens used in the Response."""
+
+    input_tokens: Optional[int] = None
+    """
+    The number of input tokens used in the Response, including text and audio
+    tokens.
+    """
+
+    output_token_details: Optional[OutputTokenDetails] = None
+    """Details about the output tokens used in the Response."""
+
+    output_tokens: Optional[int] = None
+    """
+    The number of output tokens sent in the Response, including text and audio
+    tokens.
+    """
+
+    total_tokens: Optional[int] = None
+    """
+    The total number of tokens in the Response including input and output text and
+    audio tokens.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/realtime_server_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_server_event.py
new file mode 100644
index 00000000..5f8ed55b
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/realtime_server_event.py
@@ -0,0 +1,72 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Union
+from typing_extensions import Annotated, TypeAlias
+
+from ...._utils import PropertyInfo
+from .error_event import ErrorEvent
+from .response_done_event import ResponseDoneEvent
+from .session_created_event import SessionCreatedEvent
+from .session_updated_event import SessionUpdatedEvent
+from .response_created_event import ResponseCreatedEvent
+from .response_text_done_event import ResponseTextDoneEvent
+from .rate_limits_updated_event import RateLimitsUpdatedEvent
+from .response_audio_done_event import ResponseAudioDoneEvent
+from .response_text_delta_event import ResponseTextDeltaEvent
+from .conversation_created_event import ConversationCreatedEvent
+from .response_audio_delta_event import ResponseAudioDeltaEvent
+from .conversation_item_created_event import ConversationItemCreatedEvent
+from .conversation_item_deleted_event import ConversationItemDeletedEvent
+from .response_output_item_done_event import ResponseOutputItemDoneEvent
+from .input_audio_buffer_cleared_event import InputAudioBufferClearedEvent
+from .response_content_part_done_event import ResponseContentPartDoneEvent
+from .response_output_item_added_event import ResponseOutputItemAddedEvent
+from .conversation_item_truncated_event import ConversationItemTruncatedEvent
+from .response_content_part_added_event import ResponseContentPartAddedEvent
+from .input_audio_buffer_committed_event import InputAudioBufferCommittedEvent
+from .response_audio_transcript_done_event import ResponseAudioTranscriptDoneEvent
+from .response_audio_transcript_delta_event import ResponseAudioTranscriptDeltaEvent
+from .input_audio_buffer_speech_started_event import InputAudioBufferSpeechStartedEvent
+from .input_audio_buffer_speech_stopped_event import InputAudioBufferSpeechStoppedEvent
+from .response_function_call_arguments_done_event import ResponseFunctionCallArgumentsDoneEvent
+from .response_function_call_arguments_delta_event import ResponseFunctionCallArgumentsDeltaEvent
+from .conversation_item_input_audio_transcription_failed_event import ConversationItemInputAudioTranscriptionFailedEvent
+from .conversation_item_input_audio_transcription_completed_event import (
+    ConversationItemInputAudioTranscriptionCompletedEvent,
+)
+
+__all__ = ["RealtimeServerEvent"]
+
+RealtimeServerEvent: TypeAlias = Annotated[
+    Union[
+        ErrorEvent,
+        SessionCreatedEvent,
+        SessionUpdatedEvent,
+        ConversationCreatedEvent,
+        InputAudioBufferCommittedEvent,
+        InputAudioBufferClearedEvent,
+        InputAudioBufferSpeechStartedEvent,
+        InputAudioBufferSpeechStoppedEvent,
+        ConversationItemCreatedEvent,
+        ConversationItemInputAudioTranscriptionCompletedEvent,
+        ConversationItemInputAudioTranscriptionFailedEvent,
+        ConversationItemTruncatedEvent,
+        ConversationItemDeletedEvent,
+        ResponseCreatedEvent,
+        ResponseDoneEvent,
+        ResponseOutputItemAddedEvent,
+        ResponseOutputItemDoneEvent,
+        ResponseContentPartAddedEvent,
+        ResponseContentPartDoneEvent,
+        ResponseTextDeltaEvent,
+        ResponseTextDoneEvent,
+        ResponseAudioTranscriptDeltaEvent,
+        ResponseAudioTranscriptDoneEvent,
+        ResponseAudioDeltaEvent,
+        ResponseAudioDoneEvent,
+        ResponseFunctionCallArgumentsDeltaEvent,
+        ResponseFunctionCallArgumentsDoneEvent,
+        RateLimitsUpdatedEvent,
+    ],
+    PropertyInfo(discriminator="type"),
+]
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_delta_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_delta_event.py
new file mode 100644
index 00000000..8e0128d9
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_delta_event.py
@@ -0,0 +1,30 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ResponseAudioDeltaEvent"]
+
+
+class ResponseAudioDeltaEvent(BaseModel):
+    content_index: int
+    """The index of the content part in the item's content array."""
+
+    delta: str
+    """Base64-encoded audio data delta."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the item."""
+
+    output_index: int
+    """The index of the output item in the response."""
+
+    response_id: str
+    """The ID of the response."""
+
+    type: Literal["response.audio.delta"]
+    """The event type, must be `response.audio.delta`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_done_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_done_event.py
new file mode 100644
index 00000000..68e78bc7
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_done_event.py
@@ -0,0 +1,27 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ResponseAudioDoneEvent"]
+
+
+class ResponseAudioDoneEvent(BaseModel):
+    content_index: int
+    """The index of the content part in the item's content array."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the item."""
+
+    output_index: int
+    """The index of the output item in the response."""
+
+    response_id: str
+    """The ID of the response."""
+
+    type: Literal["response.audio.done"]
+    """The event type, must be `response.audio.done`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_transcript_delta_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_transcript_delta_event.py
new file mode 100644
index 00000000..3609948d
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_transcript_delta_event.py
@@ -0,0 +1,30 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ResponseAudioTranscriptDeltaEvent"]
+
+
+class ResponseAudioTranscriptDeltaEvent(BaseModel):
+    content_index: int
+    """The index of the content part in the item's content array."""
+
+    delta: str
+    """The transcript delta."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the item."""
+
+    output_index: int
+    """The index of the output item in the response."""
+
+    response_id: str
+    """The ID of the response."""
+
+    type: Literal["response.audio_transcript.delta"]
+    """The event type, must be `response.audio_transcript.delta`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_transcript_done_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_transcript_done_event.py
new file mode 100644
index 00000000..4e4436a9
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_audio_transcript_done_event.py
@@ -0,0 +1,30 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ResponseAudioTranscriptDoneEvent"]
+
+
+class ResponseAudioTranscriptDoneEvent(BaseModel):
+    content_index: int
+    """The index of the content part in the item's content array."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the item."""
+
+    output_index: int
+    """The index of the output item in the response."""
+
+    response_id: str
+    """The ID of the response."""
+
+    transcript: str
+    """The final transcript of the audio."""
+
+    type: Literal["response.audio_transcript.done"]
+    """The event type, must be `response.audio_transcript.done`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_cancel_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_cancel_event.py
new file mode 100644
index 00000000..c5ff991e
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_cancel_event.py
@@ -0,0 +1,22 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ResponseCancelEvent"]
+
+
+class ResponseCancelEvent(BaseModel):
+    type: Literal["response.cancel"]
+    """The event type, must be `response.cancel`."""
+
+    event_id: Optional[str] = None
+    """Optional client-generated ID used to identify this event."""
+
+    response_id: Optional[str] = None
+    """
+    A specific response ID to cancel - if not provided, will cancel an in-progress
+    response in the default conversation.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_cancel_event_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_cancel_event_param.py
new file mode 100644
index 00000000..f3374073
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_cancel_event_param.py
@@ -0,0 +1,21 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["ResponseCancelEventParam"]
+
+
+class ResponseCancelEventParam(TypedDict, total=False):
+    type: Required[Literal["response.cancel"]]
+    """The event type, must be `response.cancel`."""
+
+    event_id: str
+    """Optional client-generated ID used to identify this event."""
+
+    response_id: str
+    """
+    A specific response ID to cancel - if not provided, will cancel an in-progress
+    response in the default conversation.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_content_part_added_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_content_part_added_event.py
new file mode 100644
index 00000000..45c8f20f
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_content_part_added_event.py
@@ -0,0 +1,45 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ResponseContentPartAddedEvent", "Part"]
+
+
+class Part(BaseModel):
+    audio: Optional[str] = None
+    """Base64-encoded audio data (if type is "audio")."""
+
+    text: Optional[str] = None
+    """The text content (if type is "text")."""
+
+    transcript: Optional[str] = None
+    """The transcript of the audio (if type is "audio")."""
+
+    type: Optional[Literal["text", "audio"]] = None
+    """The content type ("text", "audio")."""
+
+
+class ResponseContentPartAddedEvent(BaseModel):
+    content_index: int
+    """The index of the content part in the item's content array."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the item to which the content part was added."""
+
+    output_index: int
+    """The index of the output item in the response."""
+
+    part: Part
+    """The content part that was added."""
+
+    response_id: str
+    """The ID of the response."""
+
+    type: Literal["response.content_part.added"]
+    """The event type, must be `response.content_part.added`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_content_part_done_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_content_part_done_event.py
new file mode 100644
index 00000000..3d161161
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_content_part_done_event.py
@@ -0,0 +1,45 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ResponseContentPartDoneEvent", "Part"]
+
+
+class Part(BaseModel):
+    audio: Optional[str] = None
+    """Base64-encoded audio data (if type is "audio")."""
+
+    text: Optional[str] = None
+    """The text content (if type is "text")."""
+
+    transcript: Optional[str] = None
+    """The transcript of the audio (if type is "audio")."""
+
+    type: Optional[Literal["text", "audio"]] = None
+    """The content type ("text", "audio")."""
+
+
+class ResponseContentPartDoneEvent(BaseModel):
+    content_index: int
+    """The index of the content part in the item's content array."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the item."""
+
+    output_index: int
+    """The index of the output item in the response."""
+
+    part: Part
+    """The content part that is done."""
+
+    response_id: str
+    """The ID of the response."""
+
+    type: Literal["response.content_part.done"]
+    """The event type, must be `response.content_part.done`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_create_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_create_event.py
new file mode 100644
index 00000000..00ba1e5d
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_create_event.py
@@ -0,0 +1,115 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List, Union, Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+from .conversation_item import ConversationItem
+
+__all__ = ["ResponseCreateEvent", "Response", "ResponseTool"]
+
+
+class ResponseTool(BaseModel):
+    description: Optional[str] = None
+    """
+    The description of the function, including guidance on when and how to call it,
+    and guidance about what to tell the user when calling (if anything).
+    """
+
+    name: Optional[str] = None
+    """The name of the function."""
+
+    parameters: Optional[object] = None
+    """Parameters of the function in JSON Schema."""
+
+    type: Optional[Literal["function"]] = None
+    """The type of the tool, i.e. `function`."""
+
+
+class Response(BaseModel):
+    conversation: Union[str, Literal["auto", "none"], None] = None
+    """Controls which conversation the response is added to.
+
+    Currently supports `auto` and `none`, with `auto` as the default value. The
+    `auto` value means that the contents of the response will be added to the
+    default conversation. Set this to `none` to create an out-of-band response which
+    will not add items to default conversation.
+    """
+
+    input: Optional[List[ConversationItem]] = None
+    """Input items to include in the prompt for the model.
+
+    Creates a new context for this response, without including the default
+    conversation. Can include references to items from the default conversation.
+    """
+
+    instructions: Optional[str] = None
+    """The default system instructions (i.e.
+
+    system message) prepended to model calls. This field allows the client to guide
+    the model on desired responses. The model can be instructed on response content
+    and format, (e.g. "be extremely succinct", "act friendly", "here are examples of
+    good responses") and on audio behavior (e.g. "talk quickly", "inject emotion
+    into your voice", "laugh frequently"). The instructions are not guaranteed to be
+    followed by the model, but they provide guidance to the model on the desired
+    behavior.
+
+    Note that the server sets default instructions which will be used if this field
+    is not set and are visible in the `session.created` event at the start of the
+    session.
+    """
+
+    max_response_output_tokens: Union[int, Literal["inf"], None] = None
+    """
+    Maximum number of output tokens for a single assistant response, inclusive of
+    tool calls. Provide an integer between 1 and 4096 to limit output tokens, or
+    `inf` for the maximum available tokens for a given model. Defaults to `inf`.
+    """
+
+    metadata: Optional[object] = None
+    """Set of 16 key-value pairs that can be attached to an object.
+
+    This can be useful for storing additional information about the object in a
+    structured format. Keys can be a maximum of 64 characters long and values can be
+    a maximum of 512 characters long.
+    """
+
+    modalities: Optional[List[Literal["text", "audio"]]] = None
+    """The set of modalities the model can respond with.
+
+    To disable audio, set this to ["text"].
+    """
+
+    output_audio_format: Optional[Literal["pcm16", "g711_ulaw", "g711_alaw"]] = None
+    """The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    temperature: Optional[float] = None
+    """Sampling temperature for the model, limited to [0.6, 1.2]. Defaults to 0.8."""
+
+    tool_choice: Optional[str] = None
+    """How the model chooses tools.
+
+    Options are `auto`, `none`, `required`, or specify a function.
+    """
+
+    tools: Optional[List[ResponseTool]] = None
+    """Tools (functions) available to the model."""
+
+    voice: Optional[Literal["alloy", "ash", "ballad", "coral", "echo", "sage", "shimmer", "verse"]] = None
+    """The voice the model uses to respond.
+
+    Voice cannot be changed during the session once the model has responded with
+    audio at least once. Current voice options are `alloy`, `ash`, `ballad`,
+    `coral`, `echo` `sage`, `shimmer` and `verse`.
+    """
+
+
+class ResponseCreateEvent(BaseModel):
+    type: Literal["response.create"]
+    """The event type, must be `response.create`."""
+
+    event_id: Optional[str] = None
+    """Optional client-generated ID used to identify this event."""
+
+    response: Optional[Response] = None
+    """Create a new Realtime response with these parameters"""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_create_event_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_create_event_param.py
new file mode 100644
index 00000000..7c92b32d
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_create_event_param.py
@@ -0,0 +1,116 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import List, Union, Iterable, Optional
+from typing_extensions import Literal, Required, TypedDict
+
+from .conversation_item_param import ConversationItemParam
+
+__all__ = ["ResponseCreateEventParam", "Response", "ResponseTool"]
+
+
+class ResponseTool(TypedDict, total=False):
+    description: str
+    """
+    The description of the function, including guidance on when and how to call it,
+    and guidance about what to tell the user when calling (if anything).
+    """
+
+    name: str
+    """The name of the function."""
+
+    parameters: object
+    """Parameters of the function in JSON Schema."""
+
+    type: Literal["function"]
+    """The type of the tool, i.e. `function`."""
+
+
+class Response(TypedDict, total=False):
+    conversation: Union[str, Literal["auto", "none"]]
+    """Controls which conversation the response is added to.
+
+    Currently supports `auto` and `none`, with `auto` as the default value. The
+    `auto` value means that the contents of the response will be added to the
+    default conversation. Set this to `none` to create an out-of-band response which
+    will not add items to default conversation.
+    """
+
+    input: Iterable[ConversationItemParam]
+    """Input items to include in the prompt for the model.
+
+    Creates a new context for this response, without including the default
+    conversation. Can include references to items from the default conversation.
+    """
+
+    instructions: str
+    """The default system instructions (i.e.
+
+    system message) prepended to model calls. This field allows the client to guide
+    the model on desired responses. The model can be instructed on response content
+    and format, (e.g. "be extremely succinct", "act friendly", "here are examples of
+    good responses") and on audio behavior (e.g. "talk quickly", "inject emotion
+    into your voice", "laugh frequently"). The instructions are not guaranteed to be
+    followed by the model, but they provide guidance to the model on the desired
+    behavior.
+
+    Note that the server sets default instructions which will be used if this field
+    is not set and are visible in the `session.created` event at the start of the
+    session.
+    """
+
+    max_response_output_tokens: Union[int, Literal["inf"]]
+    """
+    Maximum number of output tokens for a single assistant response, inclusive of
+    tool calls. Provide an integer between 1 and 4096 to limit output tokens, or
+    `inf` for the maximum available tokens for a given model. Defaults to `inf`.
+    """
+
+    metadata: Optional[object]
+    """Set of 16 key-value pairs that can be attached to an object.
+
+    This can be useful for storing additional information about the object in a
+    structured format. Keys can be a maximum of 64 characters long and values can be
+    a maximum of 512 characters long.
+    """
+
+    modalities: List[Literal["text", "audio"]]
+    """The set of modalities the model can respond with.
+
+    To disable audio, set this to ["text"].
+    """
+
+    output_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"]
+    """The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    temperature: float
+    """Sampling temperature for the model, limited to [0.6, 1.2]. Defaults to 0.8."""
+
+    tool_choice: str
+    """How the model chooses tools.
+
+    Options are `auto`, `none`, `required`, or specify a function.
+    """
+
+    tools: Iterable[ResponseTool]
+    """Tools (functions) available to the model."""
+
+    voice: Literal["alloy", "ash", "ballad", "coral", "echo", "sage", "shimmer", "verse"]
+    """The voice the model uses to respond.
+
+    Voice cannot be changed during the session once the model has responded with
+    audio at least once. Current voice options are `alloy`, `ash`, `ballad`,
+    `coral`, `echo` `sage`, `shimmer` and `verse`.
+    """
+
+
+class ResponseCreateEventParam(TypedDict, total=False):
+    type: Required[Literal["response.create"]]
+    """The event type, must be `response.create`."""
+
+    event_id: str
+    """Optional client-generated ID used to identify this event."""
+
+    response: Response
+    """Create a new Realtime response with these parameters"""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_created_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_created_event.py
new file mode 100644
index 00000000..a4990cf0
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_created_event.py
@@ -0,0 +1,19 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+from .realtime_response import RealtimeResponse
+
+__all__ = ["ResponseCreatedEvent"]
+
+
+class ResponseCreatedEvent(BaseModel):
+    event_id: str
+    """The unique ID of the server event."""
+
+    response: RealtimeResponse
+    """The response resource."""
+
+    type: Literal["response.created"]
+    """The event type, must be `response.created`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_done_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_done_event.py
new file mode 100644
index 00000000..9e655184
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_done_event.py
@@ -0,0 +1,19 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+from .realtime_response import RealtimeResponse
+
+__all__ = ["ResponseDoneEvent"]
+
+
+class ResponseDoneEvent(BaseModel):
+    event_id: str
+    """The unique ID of the server event."""
+
+    response: RealtimeResponse
+    """The response resource."""
+
+    type: Literal["response.done"]
+    """The event type, must be `response.done`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_function_call_arguments_delta_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_function_call_arguments_delta_event.py
new file mode 100644
index 00000000..cdbb64e6
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_function_call_arguments_delta_event.py
@@ -0,0 +1,30 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ResponseFunctionCallArgumentsDeltaEvent"]
+
+
+class ResponseFunctionCallArgumentsDeltaEvent(BaseModel):
+    call_id: str
+    """The ID of the function call."""
+
+    delta: str
+    """The arguments delta as a JSON string."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the function call item."""
+
+    output_index: int
+    """The index of the output item in the response."""
+
+    response_id: str
+    """The ID of the response."""
+
+    type: Literal["response.function_call_arguments.delta"]
+    """The event type, must be `response.function_call_arguments.delta`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_function_call_arguments_done_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_function_call_arguments_done_event.py
new file mode 100644
index 00000000..0a5db533
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_function_call_arguments_done_event.py
@@ -0,0 +1,30 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ResponseFunctionCallArgumentsDoneEvent"]
+
+
+class ResponseFunctionCallArgumentsDoneEvent(BaseModel):
+    arguments: str
+    """The final arguments as a JSON string."""
+
+    call_id: str
+    """The ID of the function call."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the function call item."""
+
+    output_index: int
+    """The index of the output item in the response."""
+
+    response_id: str
+    """The ID of the response."""
+
+    type: Literal["response.function_call_arguments.done"]
+    """The event type, must be `response.function_call_arguments.done`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_output_item_added_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_output_item_added_event.py
new file mode 100644
index 00000000..c89bfdc3
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_output_item_added_event.py
@@ -0,0 +1,25 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+from .conversation_item import ConversationItem
+
+__all__ = ["ResponseOutputItemAddedEvent"]
+
+
+class ResponseOutputItemAddedEvent(BaseModel):
+    event_id: str
+    """The unique ID of the server event."""
+
+    item: ConversationItem
+    """The item to add to the conversation."""
+
+    output_index: int
+    """The index of the output item in the Response."""
+
+    response_id: str
+    """The ID of the Response to which the item belongs."""
+
+    type: Literal["response.output_item.added"]
+    """The event type, must be `response.output_item.added`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_output_item_done_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_output_item_done_event.py
new file mode 100644
index 00000000..b5910e22
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_output_item_done_event.py
@@ -0,0 +1,25 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+from .conversation_item import ConversationItem
+
+__all__ = ["ResponseOutputItemDoneEvent"]
+
+
+class ResponseOutputItemDoneEvent(BaseModel):
+    event_id: str
+    """The unique ID of the server event."""
+
+    item: ConversationItem
+    """The item to add to the conversation."""
+
+    output_index: int
+    """The index of the output item in the Response."""
+
+    response_id: str
+    """The ID of the Response to which the item belongs."""
+
+    type: Literal["response.output_item.done"]
+    """The event type, must be `response.output_item.done`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_text_delta_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_text_delta_event.py
new file mode 100644
index 00000000..c463b3c3
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_text_delta_event.py
@@ -0,0 +1,30 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ResponseTextDeltaEvent"]
+
+
+class ResponseTextDeltaEvent(BaseModel):
+    content_index: int
+    """The index of the content part in the item's content array."""
+
+    delta: str
+    """The text delta."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the item."""
+
+    output_index: int
+    """The index of the output item in the response."""
+
+    response_id: str
+    """The ID of the response."""
+
+    type: Literal["response.text.delta"]
+    """The event type, must be `response.text.delta`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/response_text_done_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/response_text_done_event.py
new file mode 100644
index 00000000..020ff41d
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/response_text_done_event.py
@@ -0,0 +1,30 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["ResponseTextDoneEvent"]
+
+
+class ResponseTextDoneEvent(BaseModel):
+    content_index: int
+    """The index of the content part in the item's content array."""
+
+    event_id: str
+    """The unique ID of the server event."""
+
+    item_id: str
+    """The ID of the item."""
+
+    output_index: int
+    """The index of the output item in the response."""
+
+    response_id: str
+    """The ID of the response."""
+
+    text: str
+    """The final text content."""
+
+    type: Literal["response.text.done"]
+    """The event type, must be `response.text.done`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/session.py b/portkey_ai/_vendor/openai/types/beta/realtime/session.py
new file mode 100644
index 00000000..09cdbb02
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/session.py
@@ -0,0 +1,148 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List, Union, Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["Session", "InputAudioTranscription", "Tool", "TurnDetection"]
+
+
+class InputAudioTranscription(BaseModel):
+    model: Optional[str] = None
+    """
+    The model to use for transcription, `whisper-1` is the only currently supported
+    model.
+    """
+
+
+class Tool(BaseModel):
+    description: Optional[str] = None
+    """
+    The description of the function, including guidance on when and how to call it,
+    and guidance about what to tell the user when calling (if anything).
+    """
+
+    name: Optional[str] = None
+    """The name of the function."""
+
+    parameters: Optional[object] = None
+    """Parameters of the function in JSON Schema."""
+
+    type: Optional[Literal["function"]] = None
+    """The type of the tool, i.e. `function`."""
+
+
+class TurnDetection(BaseModel):
+    prefix_padding_ms: Optional[int] = None
+    """Amount of audio to include before the VAD detected speech (in milliseconds).
+
+    Defaults to 300ms.
+    """
+
+    silence_duration_ms: Optional[int] = None
+    """Duration of silence to detect speech stop (in milliseconds).
+
+    Defaults to 500ms. With shorter values the model will respond more quickly, but
+    may jump in on short pauses from the user.
+    """
+
+    threshold: Optional[float] = None
+    """Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5.
+
+    A higher threshold will require louder audio to activate the model, and thus
+    might perform better in noisy environments.
+    """
+
+    type: Optional[Literal["server_vad"]] = None
+    """Type of turn detection, only `server_vad` is currently supported."""
+
+
+class Session(BaseModel):
+    id: Optional[str] = None
+    """Unique identifier for the session object."""
+
+    input_audio_format: Optional[Literal["pcm16", "g711_ulaw", "g711_alaw"]] = None
+    """The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    input_audio_transcription: Optional[InputAudioTranscription] = None
+    """
+    Configuration for input audio transcription, defaults to off and can be set to
+    `null` to turn off once on. Input audio transcription is not native to the
+    model, since the model consumes audio directly. Transcription runs
+    asynchronously through Whisper and should be treated as rough guidance rather
+    than the representation understood by the model.
+    """
+
+    instructions: Optional[str] = None
+    """The default system instructions (i.e.
+
+    system message) prepended to model calls. This field allows the client to guide
+    the model on desired responses. The model can be instructed on response content
+    and format, (e.g. "be extremely succinct", "act friendly", "here are examples of
+    good responses") and on audio behavior (e.g. "talk quickly", "inject emotion
+    into your voice", "laugh frequently"). The instructions are not guaranteed to be
+    followed by the model, but they provide guidance to the model on the desired
+    behavior.
+
+    Note that the server sets default instructions which will be used if this field
+    is not set and are visible in the `session.created` event at the start of the
+    session.
+    """
+
+    max_response_output_tokens: Union[int, Literal["inf"], None] = None
+    """
+    Maximum number of output tokens for a single assistant response, inclusive of
+    tool calls. Provide an integer between 1 and 4096 to limit output tokens, or
+    `inf` for the maximum available tokens for a given model. Defaults to `inf`.
+    """
+
+    modalities: Optional[List[Literal["text", "audio"]]] = None
+    """The set of modalities the model can respond with.
+
+    To disable audio, set this to ["text"].
+    """
+
+    model: Union[
+        str,
+        Literal[
+            "gpt-4o-realtime-preview",
+            "gpt-4o-realtime-preview-2024-10-01",
+            "gpt-4o-realtime-preview-2024-12-17",
+            "gpt-4o-mini-realtime-preview",
+            "gpt-4o-mini-realtime-preview-2024-12-17",
+        ],
+        None,
+    ] = None
+    """The Realtime model used for this session."""
+
+    output_audio_format: Optional[Literal["pcm16", "g711_ulaw", "g711_alaw"]] = None
+    """The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    temperature: Optional[float] = None
+    """Sampling temperature for the model, limited to [0.6, 1.2]. Defaults to 0.8."""
+
+    tool_choice: Optional[str] = None
+    """How the model chooses tools.
+
+    Options are `auto`, `none`, `required`, or specify a function.
+    """
+
+    tools: Optional[List[Tool]] = None
+    """Tools (functions) available to the model."""
+
+    turn_detection: Optional[TurnDetection] = None
+    """Configuration for turn detection.
+
+    Can be set to `null` to turn off. Server VAD means that the model will detect
+    the start and end of speech based on audio volume and respond at the end of user
+    speech.
+    """
+
+    voice: Optional[Literal["alloy", "ash", "ballad", "coral", "echo", "sage", "shimmer", "verse"]] = None
+    """The voice the model uses to respond.
+
+    Voice cannot be changed during the session once the model has responded with
+    audio at least once. Current voice options are `alloy`, `ash`, `ballad`,
+    `coral`, `echo` `sage`, `shimmer` and `verse`.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/session_create_params.py b/portkey_ai/_vendor/openai/types/beta/realtime/session_create_params.py
new file mode 100644
index 00000000..f56f2c5c
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/session_create_params.py
@@ -0,0 +1,149 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import List, Union, Iterable
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["SessionCreateParams", "InputAudioTranscription", "Tool", "TurnDetection"]
+
+
+class SessionCreateParams(TypedDict, total=False):
+    model: Required[
+        Literal[
+            "gpt-4o-realtime-preview",
+            "gpt-4o-realtime-preview-2024-10-01",
+            "gpt-4o-realtime-preview-2024-12-17",
+            "gpt-4o-mini-realtime-preview",
+            "gpt-4o-mini-realtime-preview-2024-12-17",
+        ]
+    ]
+    """The Realtime model used for this session."""
+
+    input_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"]
+    """The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    input_audio_transcription: InputAudioTranscription
+    """
+    Configuration for input audio transcription, defaults to off and can be set to
+    `null` to turn off once on. Input audio transcription is not native to the
+    model, since the model consumes audio directly. Transcription runs
+    asynchronously through Whisper and should be treated as rough guidance rather
+    than the representation understood by the model.
+    """
+
+    instructions: str
+    """The default system instructions (i.e.
+
+    system message) prepended to model calls. This field allows the client to guide
+    the model on desired responses. The model can be instructed on response content
+    and format, (e.g. "be extremely succinct", "act friendly", "here are examples of
+    good responses") and on audio behavior (e.g. "talk quickly", "inject emotion
+    into your voice", "laugh frequently"). The instructions are not guaranteed to be
+    followed by the model, but they provide guidance to the model on the desired
+    behavior.
+
+    Note that the server sets default instructions which will be used if this field
+    is not set and are visible in the `session.created` event at the start of the
+    session.
+    """
+
+    max_response_output_tokens: Union[int, Literal["inf"]]
+    """
+    Maximum number of output tokens for a single assistant response, inclusive of
+    tool calls. Provide an integer between 1 and 4096 to limit output tokens, or
+    `inf` for the maximum available tokens for a given model. Defaults to `inf`.
+    """
+
+    modalities: List[Literal["text", "audio"]]
+    """The set of modalities the model can respond with.
+
+    To disable audio, set this to ["text"].
+    """
+
+    output_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"]
+    """The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    temperature: float
+    """Sampling temperature for the model, limited to [0.6, 1.2]. Defaults to 0.8."""
+
+    tool_choice: str
+    """How the model chooses tools.
+
+    Options are `auto`, `none`, `required`, or specify a function.
+    """
+
+    tools: Iterable[Tool]
+    """Tools (functions) available to the model."""
+
+    turn_detection: TurnDetection
+    """Configuration for turn detection.
+
+    Can be set to `null` to turn off. Server VAD means that the model will detect
+    the start and end of speech based on audio volume and respond at the end of user
+    speech.
+    """
+
+    voice: Literal["alloy", "ash", "ballad", "coral", "echo", "sage", "shimmer", "verse"]
+    """The voice the model uses to respond.
+
+    Voice cannot be changed during the session once the model has responded with
+    audio at least once. Current voice options are `alloy`, `ash`, `ballad`,
+    `coral`, `echo` `sage`, `shimmer` and `verse`.
+    """
+
+
+class InputAudioTranscription(TypedDict, total=False):
+    model: str
+    """
+    The model to use for transcription, `whisper-1` is the only currently supported
+    model.
+    """
+
+
+class Tool(TypedDict, total=False):
+    description: str
+    """
+    The description of the function, including guidance on when and how to call it,
+    and guidance about what to tell the user when calling (if anything).
+    """
+
+    name: str
+    """The name of the function."""
+
+    parameters: object
+    """Parameters of the function in JSON Schema."""
+
+    type: Literal["function"]
+    """The type of the tool, i.e. `function`."""
+
+
+class TurnDetection(TypedDict, total=False):
+    create_response: bool
+    """Whether or not to automatically generate a response when VAD is enabled.
+
+    `true` by default.
+    """
+
+    prefix_padding_ms: int
+    """Amount of audio to include before the VAD detected speech (in milliseconds).
+
+    Defaults to 300ms.
+    """
+
+    silence_duration_ms: int
+    """Duration of silence to detect speech stop (in milliseconds).
+
+    Defaults to 500ms. With shorter values the model will respond more quickly, but
+    may jump in on short pauses from the user.
+    """
+
+    threshold: float
+    """Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5.
+
+    A higher threshold will require louder audio to activate the model, and thus
+    might perform better in noisy environments.
+    """
+
+    type: str
+    """Type of turn detection, only `server_vad` is currently supported."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/session_create_response.py b/portkey_ai/_vendor/openai/types/beta/realtime/session_create_response.py
new file mode 100644
index 00000000..31f591b2
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/session_create_response.py
@@ -0,0 +1,150 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List, Union, Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["SessionCreateResponse", "ClientSecret", "InputAudioTranscription", "Tool", "TurnDetection"]
+
+
+class ClientSecret(BaseModel):
+    expires_at: Optional[int] = None
+    """Timestamp for when the token expires.
+
+    Currently, all tokens expire after one minute.
+    """
+
+    value: Optional[str] = None
+    """
+    Ephemeral key usable in client environments to authenticate connections to the
+    Realtime API. Use this in client-side environments rather than a standard API
+    token, which should only be used server-side.
+    """
+
+
+class InputAudioTranscription(BaseModel):
+    model: Optional[str] = None
+    """
+    The model to use for transcription, `whisper-1` is the only currently supported
+    model.
+    """
+
+
+class Tool(BaseModel):
+    description: Optional[str] = None
+    """
+    The description of the function, including guidance on when and how to call it,
+    and guidance about what to tell the user when calling (if anything).
+    """
+
+    name: Optional[str] = None
+    """The name of the function."""
+
+    parameters: Optional[object] = None
+    """Parameters of the function in JSON Schema."""
+
+    type: Optional[Literal["function"]] = None
+    """The type of the tool, i.e. `function`."""
+
+
+class TurnDetection(BaseModel):
+    prefix_padding_ms: Optional[int] = None
+    """Amount of audio to include before the VAD detected speech (in milliseconds).
+
+    Defaults to 300ms.
+    """
+
+    silence_duration_ms: Optional[int] = None
+    """Duration of silence to detect speech stop (in milliseconds).
+
+    Defaults to 500ms. With shorter values the model will respond more quickly, but
+    may jump in on short pauses from the user.
+    """
+
+    threshold: Optional[float] = None
+    """Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5.
+
+    A higher threshold will require louder audio to activate the model, and thus
+    might perform better in noisy environments.
+    """
+
+    type: Optional[str] = None
+    """Type of turn detection, only `server_vad` is currently supported."""
+
+
+class SessionCreateResponse(BaseModel):
+    client_secret: Optional[ClientSecret] = None
+    """Ephemeral key returned by the API."""
+
+    input_audio_format: Optional[str] = None
+    """The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    input_audio_transcription: Optional[InputAudioTranscription] = None
+    """
+    Configuration for input audio transcription, defaults to off and can be set to
+    `null` to turn off once on. Input audio transcription is not native to the
+    model, since the model consumes audio directly. Transcription runs
+    asynchronously through Whisper and should be treated as rough guidance rather
+    than the representation understood by the model.
+    """
+
+    instructions: Optional[str] = None
+    """The default system instructions (i.e.
+
+    system message) prepended to model calls. This field allows the client to guide
+    the model on desired responses. The model can be instructed on response content
+    and format, (e.g. "be extremely succinct", "act friendly", "here are examples of
+    good responses") and on audio behavior (e.g. "talk quickly", "inject emotion
+    into your voice", "laugh frequently"). The instructions are not guaranteed to be
+    followed by the model, but they provide guidance to the model on the desired
+    behavior.
+
+    Note that the server sets default instructions which will be used if this field
+    is not set and are visible in the `session.created` event at the start of the
+    session.
+    """
+
+    max_response_output_tokens: Union[int, Literal["inf"], None] = None
+    """
+    Maximum number of output tokens for a single assistant response, inclusive of
+    tool calls. Provide an integer between 1 and 4096 to limit output tokens, or
+    `inf` for the maximum available tokens for a given model. Defaults to `inf`.
+    """
+
+    modalities: Optional[List[Literal["text", "audio"]]] = None
+    """The set of modalities the model can respond with.
+
+    To disable audio, set this to ["text"].
+    """
+
+    output_audio_format: Optional[str] = None
+    """The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    temperature: Optional[float] = None
+    """Sampling temperature for the model, limited to [0.6, 1.2]. Defaults to 0.8."""
+
+    tool_choice: Optional[str] = None
+    """How the model chooses tools.
+
+    Options are `auto`, `none`, `required`, or specify a function.
+    """
+
+    tools: Optional[List[Tool]] = None
+    """Tools (functions) available to the model."""
+
+    turn_detection: Optional[TurnDetection] = None
+    """Configuration for turn detection.
+
+    Can be set to `null` to turn off. Server VAD means that the model will detect
+    the start and end of speech based on audio volume and respond at the end of user
+    speech.
+    """
+
+    voice: Optional[Literal["alloy", "ash", "ballad", "coral", "echo", "sage", "shimmer", "verse"]] = None
+    """The voice the model uses to respond.
+
+    Voice cannot be changed during the session once the model has responded with
+    audio at least once. Current voice options are `alloy`, `ash`, `ballad`,
+    `coral`, `echo` `sage`, `shimmer` and `verse`.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/session_created_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/session_created_event.py
new file mode 100644
index 00000000..baf6af38
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/session_created_event.py
@@ -0,0 +1,19 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from .session import Session
+from ...._models import BaseModel
+
+__all__ = ["SessionCreatedEvent"]
+
+
+class SessionCreatedEvent(BaseModel):
+    event_id: str
+    """The unique ID of the server event."""
+
+    session: Session
+    """Realtime session object configuration."""
+
+    type: Literal["session.created"]
+    """The event type, must be `session.created`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/session_update_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/session_update_event.py
new file mode 100644
index 00000000..c04220aa
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/session_update_event.py
@@ -0,0 +1,158 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List, Union, Optional
+from typing_extensions import Literal
+
+from ...._models import BaseModel
+
+__all__ = ["SessionUpdateEvent", "Session", "SessionInputAudioTranscription", "SessionTool", "SessionTurnDetection"]
+
+
+class SessionInputAudioTranscription(BaseModel):
+    model: Optional[str] = None
+    """
+    The model to use for transcription, `whisper-1` is the only currently supported
+    model.
+    """
+
+
+class SessionTool(BaseModel):
+    description: Optional[str] = None
+    """
+    The description of the function, including guidance on when and how to call it,
+    and guidance about what to tell the user when calling (if anything).
+    """
+
+    name: Optional[str] = None
+    """The name of the function."""
+
+    parameters: Optional[object] = None
+    """Parameters of the function in JSON Schema."""
+
+    type: Optional[Literal["function"]] = None
+    """The type of the tool, i.e. `function`."""
+
+
+class SessionTurnDetection(BaseModel):
+    create_response: Optional[bool] = None
+    """Whether or not to automatically generate a response when VAD is enabled.
+
+    `true` by default.
+    """
+
+    prefix_padding_ms: Optional[int] = None
+    """Amount of audio to include before the VAD detected speech (in milliseconds).
+
+    Defaults to 300ms.
+    """
+
+    silence_duration_ms: Optional[int] = None
+    """Duration of silence to detect speech stop (in milliseconds).
+
+    Defaults to 500ms. With shorter values the model will respond more quickly, but
+    may jump in on short pauses from the user.
+    """
+
+    threshold: Optional[float] = None
+    """Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5.
+
+    A higher threshold will require louder audio to activate the model, and thus
+    might perform better in noisy environments.
+    """
+
+    type: Optional[str] = None
+    """Type of turn detection, only `server_vad` is currently supported."""
+
+
+class Session(BaseModel):
+    model: Literal[
+        "gpt-4o-realtime-preview",
+        "gpt-4o-realtime-preview-2024-10-01",
+        "gpt-4o-realtime-preview-2024-12-17",
+        "gpt-4o-mini-realtime-preview",
+        "gpt-4o-mini-realtime-preview-2024-12-17",
+    ]
+    """The Realtime model used for this session."""
+
+    input_audio_format: Optional[Literal["pcm16", "g711_ulaw", "g711_alaw"]] = None
+    """The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    input_audio_transcription: Optional[SessionInputAudioTranscription] = None
+    """
+    Configuration for input audio transcription, defaults to off and can be set to
+    `null` to turn off once on. Input audio transcription is not native to the
+    model, since the model consumes audio directly. Transcription runs
+    asynchronously through Whisper and should be treated as rough guidance rather
+    than the representation understood by the model.
+    """
+
+    instructions: Optional[str] = None
+    """The default system instructions (i.e.
+
+    system message) prepended to model calls. This field allows the client to guide
+    the model on desired responses. The model can be instructed on response content
+    and format, (e.g. "be extremely succinct", "act friendly", "here are examples of
+    good responses") and on audio behavior (e.g. "talk quickly", "inject emotion
+    into your voice", "laugh frequently"). The instructions are not guaranteed to be
+    followed by the model, but they provide guidance to the model on the desired
+    behavior.
+
+    Note that the server sets default instructions which will be used if this field
+    is not set and are visible in the `session.created` event at the start of the
+    session.
+    """
+
+    max_response_output_tokens: Union[int, Literal["inf"], None] = None
+    """
+    Maximum number of output tokens for a single assistant response, inclusive of
+    tool calls. Provide an integer between 1 and 4096 to limit output tokens, or
+    `inf` for the maximum available tokens for a given model. Defaults to `inf`.
+    """
+
+    modalities: Optional[List[Literal["text", "audio"]]] = None
+    """The set of modalities the model can respond with.
+
+    To disable audio, set this to ["text"].
+    """
+
+    output_audio_format: Optional[Literal["pcm16", "g711_ulaw", "g711_alaw"]] = None
+    """The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    temperature: Optional[float] = None
+    """Sampling temperature for the model, limited to [0.6, 1.2]. Defaults to 0.8."""
+
+    tool_choice: Optional[str] = None
+    """How the model chooses tools.
+
+    Options are `auto`, `none`, `required`, or specify a function.
+    """
+
+    tools: Optional[List[SessionTool]] = None
+    """Tools (functions) available to the model."""
+
+    turn_detection: Optional[SessionTurnDetection] = None
+    """Configuration for turn detection.
+
+    Can be set to `null` to turn off. Server VAD means that the model will detect
+    the start and end of speech based on audio volume and respond at the end of user
+    speech.
+    """
+
+    voice: Optional[Literal["alloy", "ash", "ballad", "coral", "echo", "sage", "shimmer", "verse"]] = None
+    """The voice the model uses to respond.
+
+    Voice cannot be changed during the session once the model has responded with
+    audio at least once. Current voice options are `alloy`, `ash`, `ballad`,
+    `coral`, `echo` `sage`, `shimmer` and `verse`.
+    """
+
+
+class SessionUpdateEvent(BaseModel):
+    session: Session
+    """Realtime session object configuration."""
+
+    type: Literal["session.update"]
+    """The event type, must be `session.update`."""
+
+    event_id: Optional[str] = None
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/session_update_event_param.py b/portkey_ai/_vendor/openai/types/beta/realtime/session_update_event_param.py
new file mode 100644
index 00000000..aa06069b
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/session_update_event_param.py
@@ -0,0 +1,166 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import List, Union, Iterable
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = [
+    "SessionUpdateEventParam",
+    "Session",
+    "SessionInputAudioTranscription",
+    "SessionTool",
+    "SessionTurnDetection",
+]
+
+
+class SessionInputAudioTranscription(TypedDict, total=False):
+    model: str
+    """
+    The model to use for transcription, `whisper-1` is the only currently supported
+    model.
+    """
+
+
+class SessionTool(TypedDict, total=False):
+    description: str
+    """
+    The description of the function, including guidance on when and how to call it,
+    and guidance about what to tell the user when calling (if anything).
+    """
+
+    name: str
+    """The name of the function."""
+
+    parameters: object
+    """Parameters of the function in JSON Schema."""
+
+    type: Literal["function"]
+    """The type of the tool, i.e. `function`."""
+
+
+class SessionTurnDetection(TypedDict, total=False):
+    create_response: bool
+    """Whether or not to automatically generate a response when VAD is enabled.
+
+    `true` by default.
+    """
+
+    prefix_padding_ms: int
+    """Amount of audio to include before the VAD detected speech (in milliseconds).
+
+    Defaults to 300ms.
+    """
+
+    silence_duration_ms: int
+    """Duration of silence to detect speech stop (in milliseconds).
+
+    Defaults to 500ms. With shorter values the model will respond more quickly, but
+    may jump in on short pauses from the user.
+    """
+
+    threshold: float
+    """Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5.
+
+    A higher threshold will require louder audio to activate the model, and thus
+    might perform better in noisy environments.
+    """
+
+    type: str
+    """Type of turn detection, only `server_vad` is currently supported."""
+
+
+class Session(TypedDict, total=False):
+    model: Required[
+        Literal[
+            "gpt-4o-realtime-preview",
+            "gpt-4o-realtime-preview-2024-10-01",
+            "gpt-4o-realtime-preview-2024-12-17",
+            "gpt-4o-mini-realtime-preview",
+            "gpt-4o-mini-realtime-preview-2024-12-17",
+        ]
+    ]
+    """The Realtime model used for this session."""
+
+    input_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"]
+    """The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    input_audio_transcription: SessionInputAudioTranscription
+    """
+    Configuration for input audio transcription, defaults to off and can be set to
+    `null` to turn off once on. Input audio transcription is not native to the
+    model, since the model consumes audio directly. Transcription runs
+    asynchronously through Whisper and should be treated as rough guidance rather
+    than the representation understood by the model.
+    """
+
+    instructions: str
+    """The default system instructions (i.e.
+
+    system message) prepended to model calls. This field allows the client to guide
+    the model on desired responses. The model can be instructed on response content
+    and format, (e.g. "be extremely succinct", "act friendly", "here are examples of
+    good responses") and on audio behavior (e.g. "talk quickly", "inject emotion
+    into your voice", "laugh frequently"). The instructions are not guaranteed to be
+    followed by the model, but they provide guidance to the model on the desired
+    behavior.
+
+    Note that the server sets default instructions which will be used if this field
+    is not set and are visible in the `session.created` event at the start of the
+    session.
+    """
+
+    max_response_output_tokens: Union[int, Literal["inf"]]
+    """
+    Maximum number of output tokens for a single assistant response, inclusive of
+    tool calls. Provide an integer between 1 and 4096 to limit output tokens, or
+    `inf` for the maximum available tokens for a given model. Defaults to `inf`.
+    """
+
+    modalities: List[Literal["text", "audio"]]
+    """The set of modalities the model can respond with.
+
+    To disable audio, set this to ["text"].
+    """
+
+    output_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"]
+    """The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`."""
+
+    temperature: float
+    """Sampling temperature for the model, limited to [0.6, 1.2]. Defaults to 0.8."""
+
+    tool_choice: str
+    """How the model chooses tools.
+
+    Options are `auto`, `none`, `required`, or specify a function.
+    """
+
+    tools: Iterable[SessionTool]
+    """Tools (functions) available to the model."""
+
+    turn_detection: SessionTurnDetection
+    """Configuration for turn detection.
+
+    Can be set to `null` to turn off. Server VAD means that the model will detect
+    the start and end of speech based on audio volume and respond at the end of user
+    speech.
+    """
+
+    voice: Literal["alloy", "ash", "ballad", "coral", "echo", "sage", "shimmer", "verse"]
+    """The voice the model uses to respond.
+
+    Voice cannot be changed during the session once the model has responded with
+    audio at least once. Current voice options are `alloy`, `ash`, `ballad`,
+    `coral`, `echo` `sage`, `shimmer` and `verse`.
+    """
+
+
+class SessionUpdateEventParam(TypedDict, total=False):
+    session: Required[Session]
+    """Realtime session object configuration."""
+
+    type: Required[Literal["session.update"]]
+    """The event type, must be `session.update`."""
+
+    event_id: str
+    """Optional client-generated ID used to identify this event."""
diff --git a/portkey_ai/_vendor/openai/types/beta/realtime/session_updated_event.py b/portkey_ai/_vendor/openai/types/beta/realtime/session_updated_event.py
new file mode 100644
index 00000000..b9b6488e
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/realtime/session_updated_event.py
@@ -0,0 +1,19 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from .session import Session
+from ...._models import BaseModel
+
+__all__ = ["SessionUpdatedEvent"]
+
+
+class SessionUpdatedEvent(BaseModel):
+    event_id: str
+    """The unique ID of the server event."""
+
+    session: Session
+    """Realtime session object configuration."""
+
+    type: Literal["session.updated"]
+    """The event type, must be `session.updated`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/static_file_chunking_strategy.py b/portkey_ai/_vendor/openai/types/beta/static_file_chunking_strategy.py
new file mode 100644
index 00000000..60800935
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/static_file_chunking_strategy.py
@@ -0,0 +1,21 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+
+from ..._models import BaseModel
+
+__all__ = ["StaticFileChunkingStrategy"]
+
+
+class StaticFileChunkingStrategy(BaseModel):
+    chunk_overlap_tokens: int
+    """The number of tokens that overlap between chunks. The default value is `400`.
+
+    Note that the overlap must not exceed half of `max_chunk_size_tokens`.
+    """
+
+    max_chunk_size_tokens: int
+    """The maximum number of tokens in each chunk.
+
+    The default value is `800`. The minimum value is `100` and the maximum value is
+    `4096`.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/static_file_chunking_strategy_object.py b/portkey_ai/_vendor/openai/types/beta/static_file_chunking_strategy_object.py
new file mode 100644
index 00000000..896c4b83
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/static_file_chunking_strategy_object.py
@@ -0,0 +1,15 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal
+
+from ..._models import BaseModel
+from .static_file_chunking_strategy import StaticFileChunkingStrategy
+
+__all__ = ["StaticFileChunkingStrategyObject"]
+
+
+class StaticFileChunkingStrategyObject(BaseModel):
+    static: StaticFileChunkingStrategy
+
+    type: Literal["static"]
+    """Always `static`."""
diff --git a/portkey_ai/_vendor/openai/types/beta/static_file_chunking_strategy_param.py b/portkey_ai/_vendor/openai/types/beta/static_file_chunking_strategy_param.py
new file mode 100644
index 00000000..f917ac56
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/static_file_chunking_strategy_param.py
@@ -0,0 +1,22 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Required, TypedDict
+
+__all__ = ["StaticFileChunkingStrategyParam"]
+
+
+class StaticFileChunkingStrategyParam(TypedDict, total=False):
+    chunk_overlap_tokens: Required[int]
+    """The number of tokens that overlap between chunks. The default value is `400`.
+
+    Note that the overlap must not exceed half of `max_chunk_size_tokens`.
+    """
+
+    max_chunk_size_tokens: Required[int]
+    """The maximum number of tokens in each chunk.
+
+    The default value is `800`. The minimum value is `100` and the maximum value is
+    `4096`.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/thread.py b/portkey_ai/_vendor/openai/types/beta/thread.py
index 6f7a6c7d..37d50ccb 100644
--- a/portkey_ai/_vendor/openai/types/beta/thread.py
+++ b/portkey_ai/_vendor/openai/types/beta/thread.py
@@ -45,7 +45,7 @@ class Thread(BaseModel):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     object: Literal["thread"]
diff --git a/portkey_ai/_vendor/openai/types/beta/thread_create_and_run_params.py b/portkey_ai/_vendor/openai/types/beta/thread_create_and_run_params.py
index 7490b25e..8310ba12 100644
--- a/portkey_ai/_vendor/openai/types/beta/thread_create_and_run_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/thread_create_and_run_params.py
@@ -9,6 +9,7 @@
 from .function_tool_param import FunctionToolParam
 from .file_search_tool_param import FileSearchToolParam
 from .code_interpreter_tool_param import CodeInterpreterToolParam
+from .file_chunking_strategy_param import FileChunkingStrategyParam
 from .assistant_tool_choice_option_param import AssistantToolChoiceOptionParam
 from .threads.message_content_part_param import MessageContentPartParam
 from .assistant_response_format_option_param import AssistantResponseFormatOptionParam
@@ -24,10 +25,6 @@
     "ThreadToolResourcesCodeInterpreter",
     "ThreadToolResourcesFileSearch",
     "ThreadToolResourcesFileSearchVectorStore",
-    "ThreadToolResourcesFileSearchVectorStoreChunkingStrategy",
-    "ThreadToolResourcesFileSearchVectorStoreChunkingStrategyAuto",
-    "ThreadToolResourcesFileSearchVectorStoreChunkingStrategyStatic",
-    "ThreadToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic",
     "ToolResources",
     "ToolResourcesCodeInterpreter",
     "ToolResourcesFileSearch",
@@ -75,7 +72,7 @@ class ThreadCreateAndRunParamsBase(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     model: Union[str, ChatModel, None]
@@ -89,23 +86,23 @@ class ThreadCreateAndRunParamsBase(TypedDict, total=False):
     parallel_tool_calls: bool
     """
     Whether to enable
-    [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+    [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
     during tool use.
     """
 
     response_format: Optional[AssistantResponseFormatOptionParam]
     """Specifies the format that the model must output.
 
-    Compatible with [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-    [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+    Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+    [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
     and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
     Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-    Outputs which guarantees the model will match your supplied JSON schema. Learn
-    more in the
+    Outputs which ensures the model will match your supplied JSON schema. Learn more
+    in the
     [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-    Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+    Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
     message the model generates is valid JSON.
 
     **Important:** when using JSON mode, you **must** also instruct the model to
@@ -205,7 +202,7 @@ class ThreadMessage(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
 
@@ -218,44 +215,12 @@ class ThreadToolResourcesCodeInterpreter(TypedDict, total=False):
     """
 
 
-class ThreadToolResourcesFileSearchVectorStoreChunkingStrategyAuto(TypedDict, total=False):
-    type: Required[Literal["auto"]]
-    """Always `auto`."""
-
-
-class ThreadToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic(TypedDict, total=False):
-    chunk_overlap_tokens: Required[int]
-    """The number of tokens that overlap between chunks. The default value is `400`.
-
-    Note that the overlap must not exceed half of `max_chunk_size_tokens`.
-    """
-
-    max_chunk_size_tokens: Required[int]
-    """The maximum number of tokens in each chunk.
-
-    The default value is `800`. The minimum value is `100` and the maximum value is
-    `4096`.
-    """
-
-
-class ThreadToolResourcesFileSearchVectorStoreChunkingStrategyStatic(TypedDict, total=False):
-    static: Required[ThreadToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic]
-
-    type: Required[Literal["static"]]
-    """Always `static`."""
-
-
-ThreadToolResourcesFileSearchVectorStoreChunkingStrategy: TypeAlias = Union[
-    ThreadToolResourcesFileSearchVectorStoreChunkingStrategyAuto,
-    ThreadToolResourcesFileSearchVectorStoreChunkingStrategyStatic,
-]
-
-
 class ThreadToolResourcesFileSearchVectorStore(TypedDict, total=False):
-    chunking_strategy: ThreadToolResourcesFileSearchVectorStoreChunkingStrategy
+    chunking_strategy: FileChunkingStrategyParam
     """The chunking strategy used to chunk the file(s).
 
-    If not set, will use the `auto` strategy.
+    If not set, will use the `auto` strategy. Only applicable if `file_ids` is
+    non-empty.
     """
 
     file_ids: List[str]
@@ -270,7 +235,7 @@ class ThreadToolResourcesFileSearchVectorStore(TypedDict, total=False):
 
     This can be useful for storing additional information about the vector store in
     a structured format. Keys can be a maximum of 64 characters long and values can
-    be a maxium of 512 characters long.
+    be a maximum of 512 characters long.
     """
 
 
@@ -310,7 +275,7 @@ class Thread(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     tool_resources: Optional[ThreadToolResources]
@@ -367,7 +332,7 @@ class TruncationStrategy(TypedDict, total=False):
     """
 
 
-class ThreadCreateAndRunParamsNonStreaming(ThreadCreateAndRunParamsBase):
+class ThreadCreateAndRunParamsNonStreaming(ThreadCreateAndRunParamsBase, total=False):
     stream: Optional[Literal[False]]
     """
     If `true`, returns a stream of events that happen during the Run as server-sent
diff --git a/portkey_ai/_vendor/openai/types/beta/thread_create_params.py b/portkey_ai/_vendor/openai/types/beta/thread_create_params.py
index f9561aa4..3ac6c7d6 100644
--- a/portkey_ai/_vendor/openai/types/beta/thread_create_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/thread_create_params.py
@@ -6,6 +6,7 @@
 from typing_extensions import Literal, Required, TypeAlias, TypedDict
 
 from .code_interpreter_tool_param import CodeInterpreterToolParam
+from .file_chunking_strategy_param import FileChunkingStrategyParam
 from .threads.message_content_part_param import MessageContentPartParam
 
 __all__ = [
@@ -18,10 +19,6 @@
     "ToolResourcesCodeInterpreter",
     "ToolResourcesFileSearch",
     "ToolResourcesFileSearchVectorStore",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategy",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategyAuto",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategyStatic",
-    "ToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic",
 ]
 
 
@@ -37,7 +34,7 @@ class ThreadCreateParams(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     tool_resources: Optional[ToolResources]
@@ -86,7 +83,7 @@ class Message(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
 
@@ -99,43 +96,12 @@ class ToolResourcesCodeInterpreter(TypedDict, total=False):
     """
 
 
-class ToolResourcesFileSearchVectorStoreChunkingStrategyAuto(TypedDict, total=False):
-    type: Required[Literal["auto"]]
-    """Always `auto`."""
-
-
-class ToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic(TypedDict, total=False):
-    chunk_overlap_tokens: Required[int]
-    """The number of tokens that overlap between chunks. The default value is `400`.
-
-    Note that the overlap must not exceed half of `max_chunk_size_tokens`.
-    """
-
-    max_chunk_size_tokens: Required[int]
-    """The maximum number of tokens in each chunk.
-
-    The default value is `800`. The minimum value is `100` and the maximum value is
-    `4096`.
-    """
-
-
-class ToolResourcesFileSearchVectorStoreChunkingStrategyStatic(TypedDict, total=False):
-    static: Required[ToolResourcesFileSearchVectorStoreChunkingStrategyStaticStatic]
-
-    type: Required[Literal["static"]]
-    """Always `static`."""
-
-
-ToolResourcesFileSearchVectorStoreChunkingStrategy: TypeAlias = Union[
-    ToolResourcesFileSearchVectorStoreChunkingStrategyAuto, ToolResourcesFileSearchVectorStoreChunkingStrategyStatic
-]
-
-
 class ToolResourcesFileSearchVectorStore(TypedDict, total=False):
-    chunking_strategy: ToolResourcesFileSearchVectorStoreChunkingStrategy
+    chunking_strategy: FileChunkingStrategyParam
     """The chunking strategy used to chunk the file(s).
 
-    If not set, will use the `auto` strategy.
+    If not set, will use the `auto` strategy. Only applicable if `file_ids` is
+    non-empty.
     """
 
     file_ids: List[str]
@@ -150,7 +116,7 @@ class ToolResourcesFileSearchVectorStore(TypedDict, total=False):
 
     This can be useful for storing additional information about the vector store in
     a structured format. Keys can be a maximum of 64 characters long and values can
-    be a maxium of 512 characters long.
+    be a maximum of 512 characters long.
     """
 
 
diff --git a/portkey_ai/_vendor/openai/types/beta/thread_update_params.py b/portkey_ai/_vendor/openai/types/beta/thread_update_params.py
index 7210ab77..78c5ec4f 100644
--- a/portkey_ai/_vendor/openai/types/beta/thread_update_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/thread_update_params.py
@@ -14,7 +14,7 @@ class ThreadUpdateParams(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     tool_resources: Optional[ToolResources]
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/message.py b/portkey_ai/_vendor/openai/types/beta/threads/message.py
index 298a1d42..63c5c480 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/message.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/message.py
@@ -71,7 +71,7 @@ class Message(BaseModel):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     object: Literal["thread.message"]
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/message_create_params.py b/portkey_ai/_vendor/openai/types/beta/threads/message_create_params.py
index 2b450deb..2c4edfdf 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/message_create_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/message_create_params.py
@@ -32,7 +32,7 @@ class MessageCreateParams(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
 
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/message_list_params.py b/portkey_ai/_vendor/openai/types/beta/threads/message_list_params.py
index 18c2442f..a7c22a66 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/message_list_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/message_list_params.py
@@ -21,7 +21,7 @@ class MessageListParams(TypedDict, total=False):
     """A cursor for use in pagination.
 
     `before` is an object ID that defines your place in the list. For instance, if
-    you make a list request and receive 100 objects, ending with obj_foo, your
+    you make a list request and receive 100 objects, starting with obj_foo, your
     subsequent call can include before=obj_foo in order to fetch the previous page
     of the list.
     """
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/message_update_params.py b/portkey_ai/_vendor/openai/types/beta/threads/message_update_params.py
index 7000f331..e8f8cc91 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/message_update_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/message_update_params.py
@@ -16,5 +16,5 @@ class MessageUpdateParams(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/run.py b/portkey_ai/_vendor/openai/types/beta/threads/run.py
index 0579e229..ad32135b 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/run.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/run.py
@@ -138,7 +138,7 @@ class Run(BaseModel):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     model: str
@@ -154,7 +154,7 @@ class Run(BaseModel):
     parallel_tool_calls: bool
     """
     Whether to enable
-    [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+    [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
     during tool use.
     """
 
@@ -167,16 +167,16 @@ class Run(BaseModel):
     response_format: Optional[AssistantResponseFormatOption] = None
     """Specifies the format that the model must output.
 
-    Compatible with [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-    [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+    Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+    [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
     and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
     Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-    Outputs which guarantees the model will match your supplied JSON schema. Learn
-    more in the
+    Outputs which ensures the model will match your supplied JSON schema. Learn more
+    in the
     [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-    Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+    Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
     message the model generates is valid JSON.
 
     **Important:** when using JSON mode, you **must** also instruct the model to
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/run_create_params.py b/portkey_ai/_vendor/openai/types/beta/threads/run_create_params.py
index d3e6d9c4..88dc3964 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/run_create_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/run_create_params.py
@@ -2,11 +2,12 @@
 
 from __future__ import annotations
 
-from typing import Union, Iterable, Optional
+from typing import List, Union, Iterable, Optional
 from typing_extensions import Literal, Required, TypeAlias, TypedDict
 
 from ...chat_model import ChatModel
 from ..assistant_tool_param import AssistantToolParam
+from .runs.run_step_include import RunStepInclude
 from .message_content_part_param import MessageContentPartParam
 from ..code_interpreter_tool_param import CodeInterpreterToolParam
 from ..assistant_tool_choice_option_param import AssistantToolChoiceOptionParam
@@ -32,6 +33,18 @@ class RunCreateParamsBase(TypedDict, total=False):
     execute this run.
     """
 
+    include: List[RunStepInclude]
+    """A list of additional fields to include in the response.
+
+    Currently the only supported value is
+    `step_details.tool_calls[*].file_search.results[*].content` to fetch the file
+    search result content.
+
+    See the
+    [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+    for more information.
+    """
+
     additional_instructions: Optional[str]
     """Appends additional instructions at the end of the instructions for the run.
 
@@ -72,7 +85,7 @@ class RunCreateParamsBase(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     model: Union[str, ChatModel, None]
@@ -86,23 +99,23 @@ class RunCreateParamsBase(TypedDict, total=False):
     parallel_tool_calls: bool
     """
     Whether to enable
-    [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+    [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
     during tool use.
     """
 
     response_format: Optional[AssistantResponseFormatOptionParam]
     """Specifies the format that the model must output.
 
-    Compatible with [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-    [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+    Compatible with [GPT-4o](https://platform.openai.com/docs/models#gpt-4o),
+    [GPT-4 Turbo](https://platform.openai.com/docs/models#gpt-4-turbo-and-gpt-4),
     and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
 
     Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
-    Outputs which guarantees the model will match your supplied JSON schema. Learn
-    more in the
+    Outputs which ensures the model will match your supplied JSON schema. Learn more
+    in the
     [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-    Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+    Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
     message the model generates is valid JSON.
 
     **Important:** when using JSON mode, you **must** also instruct the model to
@@ -191,7 +204,7 @@ class AdditionalMessage(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
 
@@ -212,7 +225,7 @@ class TruncationStrategy(TypedDict, total=False):
     """
 
 
-class RunCreateParamsNonStreaming(RunCreateParamsBase):
+class RunCreateParamsNonStreaming(RunCreateParamsBase, total=False):
     stream: Optional[Literal[False]]
     """
     If `true`, returns a stream of events that happen during the Run as server-sent
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/run_list_params.py b/portkey_ai/_vendor/openai/types/beta/threads/run_list_params.py
index 1e32bca4..fbea54f6 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/run_list_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/run_list_params.py
@@ -21,7 +21,7 @@ class RunListParams(TypedDict, total=False):
     """A cursor for use in pagination.
 
     `before` is an object ID that defines your place in the list. For instance, if
-    you make a list request and receive 100 objects, ending with obj_foo, your
+    you make a list request and receive 100 objects, starting with obj_foo, your
     subsequent call can include before=obj_foo in order to fetch the previous page
     of the list.
     """
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/run_submit_tool_outputs_params.py b/portkey_ai/_vendor/openai/types/beta/threads/run_submit_tool_outputs_params.py
index ccb5e5e9..14772860 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/run_submit_tool_outputs_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/run_submit_tool_outputs_params.py
@@ -31,7 +31,7 @@ class ToolOutput(TypedDict, total=False):
     """
 
 
-class RunSubmitToolOutputsParamsNonStreaming(RunSubmitToolOutputsParamsBase):
+class RunSubmitToolOutputsParamsNonStreaming(RunSubmitToolOutputsParamsBase, total=False):
     stream: Optional[Literal[False]]
     """
     If `true`, returns a stream of events that happen during the Run as server-sent
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/run_update_params.py b/portkey_ai/_vendor/openai/types/beta/threads/run_update_params.py
index e595eac8..cb4f0536 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/run_update_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/run_update_params.py
@@ -16,5 +16,5 @@ class RunUpdateParams(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/runs/__init__.py b/portkey_ai/_vendor/openai/types/beta/threads/runs/__init__.py
index a312ce3d..467d5d79 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/runs/__init__.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/runs/__init__.py
@@ -6,9 +6,11 @@
 from .tool_call import ToolCall as ToolCall
 from .run_step_delta import RunStepDelta as RunStepDelta
 from .tool_call_delta import ToolCallDelta as ToolCallDelta
+from .run_step_include import RunStepInclude as RunStepInclude
 from .step_list_params import StepListParams as StepListParams
 from .function_tool_call import FunctionToolCall as FunctionToolCall
 from .run_step_delta_event import RunStepDeltaEvent as RunStepDeltaEvent
+from .step_retrieve_params import StepRetrieveParams as StepRetrieveParams
 from .code_interpreter_logs import CodeInterpreterLogs as CodeInterpreterLogs
 from .file_search_tool_call import FileSearchToolCall as FileSearchToolCall
 from .tool_call_delta_object import ToolCallDeltaObject as ToolCallDeltaObject
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/runs/file_search_tool_call.py b/portkey_ai/_vendor/openai/types/beta/threads/runs/file_search_tool_call.py
index 57c0ca9a..da4d58dc 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/runs/file_search_tool_call.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/runs/file_search_tool_call.py
@@ -1,17 +1,71 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
+from typing import List, Optional
 from typing_extensions import Literal
 
 from ....._models import BaseModel
 
-__all__ = ["FileSearchToolCall"]
+__all__ = [
+    "FileSearchToolCall",
+    "FileSearch",
+    "FileSearchRankingOptions",
+    "FileSearchResult",
+    "FileSearchResultContent",
+]
+
+
+class FileSearchRankingOptions(BaseModel):
+    ranker: Literal["default_2024_08_21"]
+    """The ranker used for the file search."""
+
+    score_threshold: float
+    """The score threshold for the file search.
+
+    All values must be a floating point number between 0 and 1.
+    """
+
+
+class FileSearchResultContent(BaseModel):
+    text: Optional[str] = None
+    """The text content of the file."""
+
+    type: Optional[Literal["text"]] = None
+    """The type of the content."""
+
+
+class FileSearchResult(BaseModel):
+    file_id: str
+    """The ID of the file that result was found in."""
+
+    file_name: str
+    """The name of the file that result was found in."""
+
+    score: float
+    """The score of the result.
+
+    All values must be a floating point number between 0 and 1.
+    """
+
+    content: Optional[List[FileSearchResultContent]] = None
+    """The content of the result that was found.
+
+    The content is only included if requested via the include query parameter.
+    """
+
+
+class FileSearch(BaseModel):
+    ranking_options: Optional[FileSearchRankingOptions] = None
+    """The ranking options for the file search."""
+
+    results: Optional[List[FileSearchResult]] = None
+    """The results of the file search."""
 
 
 class FileSearchToolCall(BaseModel):
     id: str
     """The ID of the tool call object."""
 
-    file_search: object
+    file_search: FileSearch
     """For now, this is always going to be an empty object."""
 
     type: Literal["file_search"]
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/runs/run_step.py b/portkey_ai/_vendor/openai/types/beta/threads/runs/run_step.py
index e3163c50..0445ae36 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/runs/run_step.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/runs/run_step.py
@@ -75,7 +75,7 @@ class RunStep(BaseModel):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     object: Literal["thread.run.step"]
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/runs/run_step_include.py b/portkey_ai/_vendor/openai/types/beta/threads/runs/run_step_include.py
new file mode 100644
index 00000000..8e76c1b7
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/threads/runs/run_step_include.py
@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal, TypeAlias
+
+__all__ = ["RunStepInclude"]
+
+RunStepInclude: TypeAlias = Literal["step_details.tool_calls[*].file_search.results[*].content"]
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/runs/step_list_params.py b/portkey_ai/_vendor/openai/types/beta/threads/runs/step_list_params.py
index 606d4445..a6be771d 100644
--- a/portkey_ai/_vendor/openai/types/beta/threads/runs/step_list_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/threads/runs/step_list_params.py
@@ -2,8 +2,11 @@
 
 from __future__ import annotations
 
+from typing import List
 from typing_extensions import Literal, Required, TypedDict
 
+from .run_step_include import RunStepInclude
+
 __all__ = ["StepListParams"]
 
 
@@ -23,11 +26,23 @@ class StepListParams(TypedDict, total=False):
     """A cursor for use in pagination.
 
     `before` is an object ID that defines your place in the list. For instance, if
-    you make a list request and receive 100 objects, ending with obj_foo, your
+    you make a list request and receive 100 objects, starting with obj_foo, your
     subsequent call can include before=obj_foo in order to fetch the previous page
     of the list.
     """
 
+    include: List[RunStepInclude]
+    """A list of additional fields to include in the response.
+
+    Currently the only supported value is
+    `step_details.tool_calls[*].file_search.results[*].content` to fetch the file
+    search result content.
+
+    See the
+    [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+    for more information.
+    """
+
     limit: int
     """A limit on the number of objects to be returned.
 
diff --git a/portkey_ai/_vendor/openai/types/beta/threads/runs/step_retrieve_params.py b/portkey_ai/_vendor/openai/types/beta/threads/runs/step_retrieve_params.py
new file mode 100644
index 00000000..ecbb72ed
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/beta/threads/runs/step_retrieve_params.py
@@ -0,0 +1,28 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import List
+from typing_extensions import Required, TypedDict
+
+from .run_step_include import RunStepInclude
+
+__all__ = ["StepRetrieveParams"]
+
+
+class StepRetrieveParams(TypedDict, total=False):
+    thread_id: Required[str]
+
+    run_id: Required[str]
+
+    include: List[RunStepInclude]
+    """A list of additional fields to include in the response.
+
+    Currently the only supported value is
+    `step_details.tool_calls[*].file_search.results[*].content` to fetch the file
+    search result content.
+
+    See the
+    [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search#customizing-file-search-settings)
+    for more information.
+    """
diff --git a/portkey_ai/_vendor/openai/types/beta/vector_store.py b/portkey_ai/_vendor/openai/types/beta/vector_store.py
index 488961b4..2d3ceea8 100644
--- a/portkey_ai/_vendor/openai/types/beta/vector_store.py
+++ b/portkey_ai/_vendor/openai/types/beta/vector_store.py
@@ -53,7 +53,7 @@ class VectorStore(BaseModel):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     name: str
diff --git a/portkey_ai/_vendor/openai/types/beta/vector_store_create_params.py b/portkey_ai/_vendor/openai/types/beta/vector_store_create_params.py
index 4f74af49..4fc7c389 100644
--- a/portkey_ai/_vendor/openai/types/beta/vector_store_create_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/vector_store_create_params.py
@@ -2,21 +2,16 @@
 
 from __future__ import annotations
 
-from typing import List, Union, Optional
-from typing_extensions import Literal, Required, TypeAlias, TypedDict
+from typing import List, Optional
+from typing_extensions import Literal, Required, TypedDict
 
-__all__ = [
-    "VectorStoreCreateParams",
-    "ChunkingStrategy",
-    "ChunkingStrategyAuto",
-    "ChunkingStrategyStatic",
-    "ChunkingStrategyStaticStatic",
-    "ExpiresAfter",
-]
+from .file_chunking_strategy_param import FileChunkingStrategyParam
+
+__all__ = ["VectorStoreCreateParams", "ExpiresAfter"]
 
 
 class VectorStoreCreateParams(TypedDict, total=False):
-    chunking_strategy: ChunkingStrategy
+    chunking_strategy: FileChunkingStrategyParam
     """The chunking strategy used to chunk the file(s).
 
     If not set, will use the `auto` strategy. Only applicable if `file_ids` is
@@ -38,43 +33,13 @@ class VectorStoreCreateParams(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     name: str
     """The name of the vector store."""
 
 
-class ChunkingStrategyAuto(TypedDict, total=False):
-    type: Required[Literal["auto"]]
-    """Always `auto`."""
-
-
-class ChunkingStrategyStaticStatic(TypedDict, total=False):
-    chunk_overlap_tokens: Required[int]
-    """The number of tokens that overlap between chunks. The default value is `400`.
-
-    Note that the overlap must not exceed half of `max_chunk_size_tokens`.
-    """
-
-    max_chunk_size_tokens: Required[int]
-    """The maximum number of tokens in each chunk.
-
-    The default value is `800`. The minimum value is `100` and the maximum value is
-    `4096`.
-    """
-
-
-class ChunkingStrategyStatic(TypedDict, total=False):
-    static: Required[ChunkingStrategyStaticStatic]
-
-    type: Required[Literal["static"]]
-    """Always `static`."""
-
-
-ChunkingStrategy: TypeAlias = Union[ChunkingStrategyAuto, ChunkingStrategyStatic]
-
-
 class ExpiresAfter(TypedDict, total=False):
     anchor: Required[Literal["last_active_at"]]
     """Anchor timestamp after which the expiration policy applies.
diff --git a/portkey_ai/_vendor/openai/types/beta/vector_store_list_params.py b/portkey_ai/_vendor/openai/types/beta/vector_store_list_params.py
index f39f6726..e26ff90a 100644
--- a/portkey_ai/_vendor/openai/types/beta/vector_store_list_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/vector_store_list_params.py
@@ -21,7 +21,7 @@ class VectorStoreListParams(TypedDict, total=False):
     """A cursor for use in pagination.
 
     `before` is an object ID that defines your place in the list. For instance, if
-    you make a list request and receive 100 objects, ending with obj_foo, your
+    you make a list request and receive 100 objects, starting with obj_foo, your
     subsequent call can include before=obj_foo in order to fetch the previous page
     of the list.
     """
diff --git a/portkey_ai/_vendor/openai/types/beta/vector_store_update_params.py b/portkey_ai/_vendor/openai/types/beta/vector_store_update_params.py
index 0f9593e4..ff6c068e 100644
--- a/portkey_ai/_vendor/openai/types/beta/vector_store_update_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/vector_store_update_params.py
@@ -17,7 +17,7 @@ class VectorStoreUpdateParams(TypedDict, total=False):
 
     This can be useful for storing additional information about the object in a
     structured format. Keys can be a maximum of 64 characters long and values can be
-    a maxium of 512 characters long.
+    a maximum of 512 characters long.
     """
 
     name: Optional[str]
diff --git a/portkey_ai/_vendor/openai/types/beta/vector_stores/file_batch_create_params.py b/portkey_ai/_vendor/openai/types/beta/vector_stores/file_batch_create_params.py
index e1c3303c..e42ea99c 100644
--- a/portkey_ai/_vendor/openai/types/beta/vector_stores/file_batch_create_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/vector_stores/file_batch_create_params.py
@@ -2,16 +2,12 @@
 
 from __future__ import annotations
 
-from typing import List, Union
-from typing_extensions import Literal, Required, TypeAlias, TypedDict
+from typing import List
+from typing_extensions import Required, TypedDict
 
-__all__ = [
-    "FileBatchCreateParams",
-    "ChunkingStrategy",
-    "ChunkingStrategyAutoChunkingStrategyRequestParam",
-    "ChunkingStrategyStaticChunkingStrategyRequestParam",
-    "ChunkingStrategyStaticChunkingStrategyRequestParamStatic",
-]
+from ..file_chunking_strategy_param import FileChunkingStrategyParam
+
+__all__ = ["FileBatchCreateParams"]
 
 
 class FileBatchCreateParams(TypedDict, total=False):
@@ -22,40 +18,9 @@ class FileBatchCreateParams(TypedDict, total=False):
     files.
     """
 
-    chunking_strategy: ChunkingStrategy
+    chunking_strategy: FileChunkingStrategyParam
     """The chunking strategy used to chunk the file(s).
 
-    If not set, will use the `auto` strategy.
-    """
-
-
-class ChunkingStrategyAutoChunkingStrategyRequestParam(TypedDict, total=False):
-    type: Required[Literal["auto"]]
-    """Always `auto`."""
-
-
-class ChunkingStrategyStaticChunkingStrategyRequestParamStatic(TypedDict, total=False):
-    chunk_overlap_tokens: Required[int]
-    """The number of tokens that overlap between chunks. The default value is `400`.
-
-    Note that the overlap must not exceed half of `max_chunk_size_tokens`.
+    If not set, will use the `auto` strategy. Only applicable if `file_ids` is
+    non-empty.
     """
-
-    max_chunk_size_tokens: Required[int]
-    """The maximum number of tokens in each chunk.
-
-    The default value is `800`. The minimum value is `100` and the maximum value is
-    `4096`.
-    """
-
-
-class ChunkingStrategyStaticChunkingStrategyRequestParam(TypedDict, total=False):
-    static: Required[ChunkingStrategyStaticChunkingStrategyRequestParamStatic]
-
-    type: Required[Literal["static"]]
-    """Always `static`."""
-
-
-ChunkingStrategy: TypeAlias = Union[
-    ChunkingStrategyAutoChunkingStrategyRequestParam, ChunkingStrategyStaticChunkingStrategyRequestParam
-]
diff --git a/portkey_ai/_vendor/openai/types/beta/vector_stores/file_batch_list_files_params.py b/portkey_ai/_vendor/openai/types/beta/vector_stores/file_batch_list_files_params.py
index 24dee7d5..2a0a6c6a 100644
--- a/portkey_ai/_vendor/openai/types/beta/vector_stores/file_batch_list_files_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/vector_stores/file_batch_list_files_params.py
@@ -23,7 +23,7 @@ class FileBatchListFilesParams(TypedDict, total=False):
     """A cursor for use in pagination.
 
     `before` is an object ID that defines your place in the list. For instance, if
-    you make a list request and receive 100 objects, ending with obj_foo, your
+    you make a list request and receive 100 objects, starting with obj_foo, your
     subsequent call can include before=obj_foo in order to fetch the previous page
     of the list.
     """
diff --git a/portkey_ai/_vendor/openai/types/beta/vector_stores/file_create_params.py b/portkey_ai/_vendor/openai/types/beta/vector_stores/file_create_params.py
index cfb80657..d074d766 100644
--- a/portkey_ai/_vendor/openai/types/beta/vector_stores/file_create_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/vector_stores/file_create_params.py
@@ -2,16 +2,11 @@
 
 from __future__ import annotations
 
-from typing import Union
-from typing_extensions import Literal, Required, TypeAlias, TypedDict
+from typing_extensions import Required, TypedDict
 
-__all__ = [
-    "FileCreateParams",
-    "ChunkingStrategy",
-    "ChunkingStrategyAutoChunkingStrategyRequestParam",
-    "ChunkingStrategyStaticChunkingStrategyRequestParam",
-    "ChunkingStrategyStaticChunkingStrategyRequestParamStatic",
-]
+from ..file_chunking_strategy_param import FileChunkingStrategyParam
+
+__all__ = ["FileCreateParams"]
 
 
 class FileCreateParams(TypedDict, total=False):
@@ -22,40 +17,9 @@ class FileCreateParams(TypedDict, total=False):
     files.
     """
 
-    chunking_strategy: ChunkingStrategy
+    chunking_strategy: FileChunkingStrategyParam
     """The chunking strategy used to chunk the file(s).
 
-    If not set, will use the `auto` strategy.
-    """
-
-
-class ChunkingStrategyAutoChunkingStrategyRequestParam(TypedDict, total=False):
-    type: Required[Literal["auto"]]
-    """Always `auto`."""
-
-
-class ChunkingStrategyStaticChunkingStrategyRequestParamStatic(TypedDict, total=False):
-    chunk_overlap_tokens: Required[int]
-    """The number of tokens that overlap between chunks. The default value is `400`.
-
-    Note that the overlap must not exceed half of `max_chunk_size_tokens`.
+    If not set, will use the `auto` strategy. Only applicable if `file_ids` is
+    non-empty.
     """
-
-    max_chunk_size_tokens: Required[int]
-    """The maximum number of tokens in each chunk.
-
-    The default value is `800`. The minimum value is `100` and the maximum value is
-    `4096`.
-    """
-
-
-class ChunkingStrategyStaticChunkingStrategyRequestParam(TypedDict, total=False):
-    static: Required[ChunkingStrategyStaticChunkingStrategyRequestParamStatic]
-
-    type: Required[Literal["static"]]
-    """Always `static`."""
-
-
-ChunkingStrategy: TypeAlias = Union[
-    ChunkingStrategyAutoChunkingStrategyRequestParam, ChunkingStrategyStaticChunkingStrategyRequestParam
-]
diff --git a/portkey_ai/_vendor/openai/types/beta/vector_stores/file_list_params.py b/portkey_ai/_vendor/openai/types/beta/vector_stores/file_list_params.py
index 23dd7f0d..867b5fb3 100644
--- a/portkey_ai/_vendor/openai/types/beta/vector_stores/file_list_params.py
+++ b/portkey_ai/_vendor/openai/types/beta/vector_stores/file_list_params.py
@@ -21,7 +21,7 @@ class FileListParams(TypedDict, total=False):
     """A cursor for use in pagination.
 
     `before` is an object ID that defines your place in the list. For instance, if
-    you make a list request and receive 100 objects, ending with obj_foo, your
+    you make a list request and receive 100 objects, starting with obj_foo, your
     subsequent call can include before=obj_foo in order to fetch the previous page
     of the list.
     """
diff --git a/portkey_ai/_vendor/openai/types/beta/vector_stores/vector_store_file.py b/portkey_ai/_vendor/openai/types/beta/vector_stores/vector_store_file.py
index 65096e8d..e4608e15 100644
--- a/portkey_ai/_vendor/openai/types/beta/vector_stores/vector_store_file.py
+++ b/portkey_ai/_vendor/openai/types/beta/vector_stores/vector_store_file.py
@@ -1,19 +1,12 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing import Union, Optional
-from typing_extensions import Literal, Annotated, TypeAlias
+from typing import Optional
+from typing_extensions import Literal
 
-from ...._utils import PropertyInfo
 from ...._models import BaseModel
+from ..file_chunking_strategy import FileChunkingStrategy
 
-__all__ = [
-    "VectorStoreFile",
-    "LastError",
-    "ChunkingStrategy",
-    "ChunkingStrategyStatic",
-    "ChunkingStrategyStaticStatic",
-    "ChunkingStrategyOther",
-]
+__all__ = ["VectorStoreFile", "LastError"]
 
 
 class LastError(BaseModel):
@@ -24,38 +17,6 @@ class LastError(BaseModel):
     """A human-readable description of the error."""
 
 
-class ChunkingStrategyStaticStatic(BaseModel):
-    chunk_overlap_tokens: int
-    """The number of tokens that overlap between chunks. The default value is `400`.
-
-    Note that the overlap must not exceed half of `max_chunk_size_tokens`.
-    """
-
-    max_chunk_size_tokens: int
-    """The maximum number of tokens in each chunk.
-
-    The default value is `800`. The minimum value is `100` and the maximum value is
-    `4096`.
-    """
-
-
-class ChunkingStrategyStatic(BaseModel):
-    static: ChunkingStrategyStaticStatic
-
-    type: Literal["static"]
-    """Always `static`."""
-
-
-class ChunkingStrategyOther(BaseModel):
-    type: Literal["other"]
-    """Always `other`."""
-
-
-ChunkingStrategy: TypeAlias = Annotated[
-    Union[ChunkingStrategyStatic, ChunkingStrategyOther], PropertyInfo(discriminator="type")
-]
-
-
 class VectorStoreFile(BaseModel):
     id: str
     """The identifier, which can be referenced in API endpoints."""
@@ -93,5 +54,5 @@ class VectorStoreFile(BaseModel):
     attached to.
     """
 
-    chunking_strategy: Optional[ChunkingStrategy] = None
+    chunking_strategy: Optional[FileChunkingStrategy] = None
     """The strategy used to chunk the file."""
diff --git a/portkey_ai/_vendor/openai/types/chat/__init__.py b/portkey_ai/_vendor/openai/types/chat/__init__.py
index a5cf3734..c623a982 100644
--- a/portkey_ai/_vendor/openai/types/chat/__init__.py
+++ b/portkey_ai/_vendor/openai/types/chat/__init__.py
@@ -4,6 +4,7 @@
 
 from .chat_completion import ChatCompletion as ChatCompletion
 from .chat_completion_role import ChatCompletionRole as ChatCompletionRole
+from .chat_completion_audio import ChatCompletionAudio as ChatCompletionAudio
 from .chat_completion_chunk import ChatCompletionChunk as ChatCompletionChunk
 from .parsed_chat_completion import (
     ParsedChoice as ParsedChoice,
@@ -11,14 +12,17 @@
     ParsedChatCompletionMessage as ParsedChatCompletionMessage,
 )
 from .chat_completion_message import ChatCompletionMessage as ChatCompletionMessage
+from .chat_completion_modality import ChatCompletionModality as ChatCompletionModality
 from .completion_create_params import CompletionCreateParams as CompletionCreateParams
 from .parsed_function_tool_call import (
     ParsedFunction as ParsedFunction,
     ParsedFunctionToolCall as ParsedFunctionToolCall,
 )
 from .chat_completion_tool_param import ChatCompletionToolParam as ChatCompletionToolParam
+from .chat_completion_audio_param import ChatCompletionAudioParam as ChatCompletionAudioParam
 from .chat_completion_message_param import ChatCompletionMessageParam as ChatCompletionMessageParam
 from .chat_completion_token_logprob import ChatCompletionTokenLogprob as ChatCompletionTokenLogprob
+from .chat_completion_reasoning_effort import ChatCompletionReasoningEffort as ChatCompletionReasoningEffort
 from .chat_completion_message_tool_call import ChatCompletionMessageToolCall as ChatCompletionMessageToolCall
 from .chat_completion_content_part_param import ChatCompletionContentPartParam as ChatCompletionContentPartParam
 from .chat_completion_tool_message_param import ChatCompletionToolMessageParam as ChatCompletionToolMessageParam
@@ -34,6 +38,9 @@
 from .chat_completion_content_part_text_param import (
     ChatCompletionContentPartTextParam as ChatCompletionContentPartTextParam,
 )
+from .chat_completion_developer_message_param import (
+    ChatCompletionDeveloperMessageParam as ChatCompletionDeveloperMessageParam,
+)
 from .chat_completion_message_tool_call_param import (
     ChatCompletionMessageToolCallParam as ChatCompletionMessageToolCallParam,
 )
@@ -43,6 +50,9 @@
 from .chat_completion_content_part_image_param import (
     ChatCompletionContentPartImageParam as ChatCompletionContentPartImageParam,
 )
+from .chat_completion_prediction_content_param import (
+    ChatCompletionPredictionContentParam as ChatCompletionPredictionContentParam,
+)
 from .chat_completion_tool_choice_option_param import (
     ChatCompletionToolChoiceOptionParam as ChatCompletionToolChoiceOptionParam,
 )
@@ -52,3 +62,6 @@
 from .chat_completion_function_call_option_param import (
     ChatCompletionFunctionCallOptionParam as ChatCompletionFunctionCallOptionParam,
 )
+from .chat_completion_content_part_input_audio_param import (
+    ChatCompletionContentPartInputAudioParam as ChatCompletionContentPartInputAudioParam,
+)
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_assistant_message_param.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_assistant_message_param.py
index 2429d41d..35e3a3d7 100644
--- a/portkey_ai/_vendor/openai/types/chat/chat_completion_assistant_message_param.py
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_assistant_message_param.py
@@ -9,7 +9,13 @@
 from .chat_completion_message_tool_call_param import ChatCompletionMessageToolCallParam
 from .chat_completion_content_part_refusal_param import ChatCompletionContentPartRefusalParam
 
-__all__ = ["ChatCompletionAssistantMessageParam", "ContentArrayOfContentPart", "FunctionCall"]
+__all__ = ["ChatCompletionAssistantMessageParam", "Audio", "ContentArrayOfContentPart", "FunctionCall"]
+
+
+class Audio(TypedDict, total=False):
+    id: Required[str]
+    """Unique identifier for a previous audio response from the model."""
+
 
 ContentArrayOfContentPart: TypeAlias = Union[ChatCompletionContentPartTextParam, ChatCompletionContentPartRefusalParam]
 
@@ -31,6 +37,12 @@ class ChatCompletionAssistantMessageParam(TypedDict, total=False):
     role: Required[Literal["assistant"]]
     """The role of the messages author, in this case `assistant`."""
 
+    audio: Optional[Audio]
+    """Data about a previous audio response from the model.
+
+    [Learn more](https://platform.openai.com/docs/guides/audio).
+    """
+
     content: Union[str, Iterable[ContentArrayOfContentPart], None]
     """The contents of the assistant message.
 
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_audio.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_audio.py
new file mode 100644
index 00000000..dd15508e
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_audio.py
@@ -0,0 +1,26 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+
+from ..._models import BaseModel
+
+__all__ = ["ChatCompletionAudio"]
+
+
+class ChatCompletionAudio(BaseModel):
+    id: str
+    """Unique identifier for this audio response."""
+
+    data: str
+    """
+    Base64 encoded audio bytes generated by the model, in the format specified in
+    the request.
+    """
+
+    expires_at: int
+    """
+    The Unix timestamp (in seconds) for when this audio response will no longer be
+    accessible on the server for use in multi-turn conversations.
+    """
+
+    transcript: str
+    """Transcript of the audio generated by the model."""
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_audio_param.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_audio_param.py
new file mode 100644
index 00000000..1e20a52b
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_audio_param.py
@@ -0,0 +1,23 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["ChatCompletionAudioParam"]
+
+
+class ChatCompletionAudioParam(TypedDict, total=False):
+    format: Required[Literal["wav", "mp3", "flac", "opus", "pcm16"]]
+    """Specifies the output audio format.
+
+    Must be one of `wav`, `mp3`, `flac`, `opus`, or `pcm16`.
+    """
+
+    voice: Required[Literal["alloy", "ash", "ballad", "coral", "echo", "sage", "shimmer", "verse"]]
+    """The voice the model uses to respond.
+
+    Supported voices are `ash`, `ballad`, `coral`, `sage`, and `verse` (also
+    supported but not recommended are `alloy`, `echo`, and `shimmer`; these voices
+    are less expressive).
+    """
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_content_part_image_param.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_content_part_image_param.py
index b1a186aa..9d407324 100644
--- a/portkey_ai/_vendor/openai/types/chat/chat_completion_content_part_image_param.py
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_content_part_image_param.py
@@ -15,7 +15,7 @@ class ImageURL(TypedDict, total=False):
     """Specifies the detail level of the image.
 
     Learn more in the
-    [Vision guide](https://platform.openai.com/docs/guides/vision/low-or-high-fidelity-image-understanding).
+    [Vision guide](https://platform.openai.com/docs/guides/vision#low-or-high-fidelity-image-understanding).
     """
 
 
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_content_part_input_audio_param.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_content_part_input_audio_param.py
new file mode 100644
index 00000000..0b1b1a80
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_content_part_input_audio_param.py
@@ -0,0 +1,22 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["ChatCompletionContentPartInputAudioParam", "InputAudio"]
+
+
+class InputAudio(TypedDict, total=False):
+    data: Required[str]
+    """Base64 encoded audio data."""
+
+    format: Required[Literal["wav", "mp3"]]
+    """The format of the encoded audio data. Currently supports "wav" and "mp3"."""
+
+
+class ChatCompletionContentPartInputAudioParam(TypedDict, total=False):
+    input_audio: Required[InputAudio]
+
+    type: Required[Literal["input_audio"]]
+    """The type of the content part. Always `input_audio`."""
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_content_part_param.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_content_part_param.py
index e0c6e480..682d11f4 100644
--- a/portkey_ai/_vendor/openai/types/chat/chat_completion_content_part_param.py
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_content_part_param.py
@@ -7,9 +7,10 @@
 
 from .chat_completion_content_part_text_param import ChatCompletionContentPartTextParam
 from .chat_completion_content_part_image_param import ChatCompletionContentPartImageParam
+from .chat_completion_content_part_input_audio_param import ChatCompletionContentPartInputAudioParam
 
 __all__ = ["ChatCompletionContentPartParam"]
 
 ChatCompletionContentPartParam: TypeAlias = Union[
-    ChatCompletionContentPartTextParam, ChatCompletionContentPartImageParam
+    ChatCompletionContentPartTextParam, ChatCompletionContentPartImageParam, ChatCompletionContentPartInputAudioParam
 ]
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_developer_message_param.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_developer_message_param.py
new file mode 100644
index 00000000..01e4fdb6
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_developer_message_param.py
@@ -0,0 +1,25 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union, Iterable
+from typing_extensions import Literal, Required, TypedDict
+
+from .chat_completion_content_part_text_param import ChatCompletionContentPartTextParam
+
+__all__ = ["ChatCompletionDeveloperMessageParam"]
+
+
+class ChatCompletionDeveloperMessageParam(TypedDict, total=False):
+    content: Required[Union[str, Iterable[ChatCompletionContentPartTextParam]]]
+    """The contents of the developer message."""
+
+    role: Required[Literal["developer"]]
+    """The role of the messages author, in this case `developer`."""
+
+    name: str
+    """An optional name for the participant.
+
+    Provides the model information to differentiate between participants of the same
+    role.
+    """
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_message.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_message.py
index 492bb68c..704fa5d5 100644
--- a/portkey_ai/_vendor/openai/types/chat/chat_completion_message.py
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_message.py
@@ -4,6 +4,7 @@
 from typing_extensions import Literal
 
 from ..._models import BaseModel
+from .chat_completion_audio import ChatCompletionAudio
 from .chat_completion_message_tool_call import ChatCompletionMessageToolCall
 
 __all__ = ["ChatCompletionMessage", "FunctionCall"]
@@ -32,6 +33,13 @@ class ChatCompletionMessage(BaseModel):
     role: Literal["assistant"]
     """The role of the author of this message."""
 
+    audio: Optional[ChatCompletionAudio] = None
+    """
+    If the audio output modality is requested, this object contains data about the
+    audio response from the model.
+    [Learn more](https://platform.openai.com/docs/guides/audio).
+    """
+
     function_call: Optional[FunctionCall] = None
     """Deprecated and replaced by `tool_calls`.
 
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_message_param.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_message_param.py
index ec65d94c..942da243 100644
--- a/portkey_ai/_vendor/openai/types/chat/chat_completion_message_param.py
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_message_param.py
@@ -10,10 +10,12 @@
 from .chat_completion_system_message_param import ChatCompletionSystemMessageParam
 from .chat_completion_function_message_param import ChatCompletionFunctionMessageParam
 from .chat_completion_assistant_message_param import ChatCompletionAssistantMessageParam
+from .chat_completion_developer_message_param import ChatCompletionDeveloperMessageParam
 
 __all__ = ["ChatCompletionMessageParam"]
 
 ChatCompletionMessageParam: TypeAlias = Union[
+    ChatCompletionDeveloperMessageParam,
     ChatCompletionSystemMessageParam,
     ChatCompletionUserMessageParam,
     ChatCompletionAssistantMessageParam,
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_modality.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_modality.py
new file mode 100644
index 00000000..8e3c1459
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_modality.py
@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal, TypeAlias
+
+__all__ = ["ChatCompletionModality"]
+
+ChatCompletionModality: TypeAlias = Literal["text", "audio"]
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_prediction_content_param.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_prediction_content_param.py
new file mode 100644
index 00000000..c44e6e36
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_prediction_content_param.py
@@ -0,0 +1,25 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union, Iterable
+from typing_extensions import Literal, Required, TypedDict
+
+from .chat_completion_content_part_text_param import ChatCompletionContentPartTextParam
+
+__all__ = ["ChatCompletionPredictionContentParam"]
+
+
+class ChatCompletionPredictionContentParam(TypedDict, total=False):
+    content: Required[Union[str, Iterable[ChatCompletionContentPartTextParam]]]
+    """
+    The content that should be matched when generating a model response. If
+    generated tokens would match this content, the entire model response can be
+    returned much more quickly.
+    """
+
+    type: Required[Literal["content"]]
+    """The type of the predicted content you want to provide.
+
+    This type is currently always `content`.
+    """
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_reasoning_effort.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_reasoning_effort.py
new file mode 100644
index 00000000..9e794697
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_reasoning_effort.py
@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal, TypeAlias
+
+__all__ = ["ChatCompletionReasoningEffort"]
+
+ChatCompletionReasoningEffort: TypeAlias = Literal["low", "medium", "high"]
diff --git a/portkey_ai/_vendor/openai/types/chat/chat_completion_tool_param.py b/portkey_ai/_vendor/openai/types/chat/chat_completion_tool_param.py
index 0cf6ea72..6c2b1a36 100644
--- a/portkey_ai/_vendor/openai/types/chat/chat_completion_tool_param.py
+++ b/portkey_ai/_vendor/openai/types/chat/chat_completion_tool_param.py
@@ -4,13 +4,13 @@
 
 from typing_extensions import Literal, Required, TypedDict
 
-from ...types import shared_params
+from ..shared_params.function_definition import FunctionDefinition
 
 __all__ = ["ChatCompletionToolParam"]
 
 
 class ChatCompletionToolParam(TypedDict, total=False):
-    function: Required[shared_params.FunctionDefinition]
+    function: Required[FunctionDefinition]
 
     type: Required[Literal["function"]]
     """The type of the tool. Currently, only `function` is supported."""
diff --git a/portkey_ai/_vendor/openai/types/chat/completion_create_params.py b/portkey_ai/_vendor/openai/types/chat/completion_create_params.py
index bf648a38..f168ddea 100644
--- a/portkey_ai/_vendor/openai/types/chat/completion_create_params.py
+++ b/portkey_ai/_vendor/openai/types/chat/completion_create_params.py
@@ -5,12 +5,19 @@
 from typing import Dict, List, Union, Iterable, Optional
 from typing_extensions import Literal, Required, TypeAlias, TypedDict
 
-from ...types import shared_params
 from ..chat_model import ChatModel
+from .chat_completion_modality import ChatCompletionModality
 from .chat_completion_tool_param import ChatCompletionToolParam
+from .chat_completion_audio_param import ChatCompletionAudioParam
 from .chat_completion_message_param import ChatCompletionMessageParam
+from .chat_completion_reasoning_effort import ChatCompletionReasoningEffort
+from ..shared_params.function_parameters import FunctionParameters
+from ..shared_params.response_format_text import ResponseFormatText
 from .chat_completion_stream_options_param import ChatCompletionStreamOptionsParam
+from .chat_completion_prediction_content_param import ChatCompletionPredictionContentParam
 from .chat_completion_tool_choice_option_param import ChatCompletionToolChoiceOptionParam
+from ..shared_params.response_format_json_object import ResponseFormatJSONObject
+from ..shared_params.response_format_json_schema import ResponseFormatJSONSchema
 from .chat_completion_function_call_option_param import ChatCompletionFunctionCallOptionParam
 
 __all__ = [
@@ -27,35 +34,48 @@ class CompletionCreateParamsBase(TypedDict, total=False):
     messages: Required[Iterable[ChatCompletionMessageParam]]
     """A list of messages comprising the conversation so far.
 
-    [Example Python code](https://cookbook.openai.com/examples/how_to_format_inputs_to_chatgpt_models).
+    Depending on the [model](https://platform.openai.com/docs/models) you use,
+    different message types (modalities) are supported, like
+    [text](https://platform.openai.com/docs/guides/text-generation),
+    [images](https://platform.openai.com/docs/guides/vision), and
+    [audio](https://platform.openai.com/docs/guides/audio).
     """
 
     model: Required[Union[str, ChatModel]]
     """ID of the model to use.
 
     See the
-    [model endpoint compatibility](https://platform.openai.com/docs/models/model-endpoint-compatibility)
+    [model endpoint compatibility](https://platform.openai.com/docs/models#model-endpoint-compatibility)
     table for details on which models work with the Chat API.
     """
 
+    audio: Optional[ChatCompletionAudioParam]
+    """Parameters for audio output.
+
+    Required when audio output is requested with `modalities: ["audio"]`.
+    [Learn more](https://platform.openai.com/docs/guides/audio).
+    """
+
     frequency_penalty: Optional[float]
     """Number between -2.0 and 2.0.
 
     Positive values penalize new tokens based on their existing frequency in the
     text so far, decreasing the model's likelihood to repeat the same line verbatim.
-
-    [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
     """
 
     function_call: FunctionCall
     """Deprecated in favor of `tool_choice`.
 
-    Controls which (if any) function is called by the model. `none` means the model
-    will not call a function and instead generates a message. `auto` means the model
-    can pick between generating a message or calling a function. Specifying a
-    particular function via `{"name": "my_function"}` forces the model to call that
+    Controls which (if any) function is called by the model.
+
+    `none` means the model will not call a function and instead generates a message.
+
+    `auto` means the model can pick between generating a message or calling a
     function.
 
+    Specifying a particular function via `{"name": "my_function"}` forces the model
+    to call that function.
+
     `none` is the default when no functions are present. `auto` is the default if
     functions are present.
     """
@@ -84,15 +104,42 @@ class CompletionCreateParamsBase(TypedDict, total=False):
     `content` of `message`.
     """
 
+    max_completion_tokens: Optional[int]
+    """
+    An upper bound for the number of tokens that can be generated for a completion,
+    including visible output tokens and
+    [reasoning tokens](https://platform.openai.com/docs/guides/reasoning).
+    """
+
     max_tokens: Optional[int]
     """
     The maximum number of [tokens](/tokenizer) that can be generated in the chat
-    completion.
+    completion. This value can be used to control
+    [costs](https://openai.com/api/pricing/) for text generated via API.
+
+    This value is now deprecated in favor of `max_completion_tokens`, and is not
+    compatible with
+    [o1 series models](https://platform.openai.com/docs/guides/reasoning).
+    """
+
+    metadata: Optional[Dict[str, str]]
+    """
+    Developer-defined tags and values used for filtering completions in the
+    [dashboard](https://platform.openai.com/chat-completions).
+    """
 
-    The total length of input tokens and generated tokens is limited by the model's
-    context length.
-    [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken)
-    for counting tokens.
+    modalities: Optional[List[ChatCompletionModality]]
+    """
+    Output types that you would like the model to generate for this request. Most
+    models are capable of generating text, which is the default:
+
+    `["text"]`
+
+    The `gpt-4o-audio-preview` model can also be used to
+    [generate audio](https://platform.openai.com/docs/guides/audio). To request that
+    this model generate both text and audio responses, you can use:
+
+    `["text", "audio"]`
     """
 
     n: Optional[int]
@@ -105,28 +152,41 @@ class CompletionCreateParamsBase(TypedDict, total=False):
     parallel_tool_calls: bool
     """
     Whether to enable
-    [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling)
+    [parallel function calling](https://platform.openai.com/docs/guides/function-calling#configuring-parallel-function-calling)
     during tool use.
     """
 
+    prediction: Optional[ChatCompletionPredictionContentParam]
+    """
+    Static predicted output content, such as the content of a text file that is
+    being regenerated.
+    """
+
     presence_penalty: Optional[float]
     """Number between -2.0 and 2.0.
 
     Positive values penalize new tokens based on whether they appear in the text so
     far, increasing the model's likelihood to talk about new topics.
+    """
+
+    reasoning_effort: ChatCompletionReasoningEffort
+    """**o1 models only**
 
-    [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+    Constrains effort on reasoning for
+    [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+    supported values are `low`, `medium`, and `high`. Reducing reasoning effort can
+    result in faster responses and fewer tokens used on reasoning in a response.
     """
 
     response_format: ResponseFormat
     """An object specifying the format that the model must output.
 
-    Compatible with [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
-    [GPT-4o mini](https://platform.openai.com/docs/models/gpt-4o-mini),
-    [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo) and
-    all GPT-3.5 Turbo models newer than `gpt-3.5-turbo-1106`.
+    Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
+    Outputs which ensures the model will match your supplied JSON schema. Learn more
+    in the
+    [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
 
-    Setting to `{ "type": "json_object" }` enables JSON mode, which guarantees the
+    Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
     message the model generates is valid JSON.
 
     **Important:** when using JSON mode, you **must** also instruct the model to
@@ -152,8 +212,11 @@ class CompletionCreateParamsBase(TypedDict, total=False):
 
     This parameter is relevant for customers subscribed to the scale tier service:
 
-    - If set to 'auto', the system will utilize scale tier credits until they are
-      exhausted.
+    - If set to 'auto', and the Project is Scale tier enabled, the system will
+      utilize scale tier credits until they are exhausted.
+    - If set to 'auto', and the Project is not Scale tier enabled, the request will
+      be processed using the default service tier with a lower uptime SLA and no
+      latency guarentee.
     - If set to 'default', the request will be processed using the default service
       tier with a lower uptime SLA and no latency guarentee.
     - When not set, the default behavior is 'auto'.
@@ -165,6 +228,13 @@ class CompletionCreateParamsBase(TypedDict, total=False):
     stop: Union[Optional[str], List[str]]
     """Up to 4 sequences where the API will stop generating further tokens."""
 
+    store: Optional[bool]
+    """
+    Whether or not to store the output of this chat completion request for use in
+    our [model distillation](https://platform.openai.com/docs/guides/distillation)
+    or [evals](https://platform.openai.com/docs/guides/evals) products.
+    """
+
     stream_options: Optional[ChatCompletionStreamOptionsParam]
     """Options for streaming response. Only set this when you set `stream: true`."""
 
@@ -172,9 +242,8 @@ class CompletionCreateParamsBase(TypedDict, total=False):
     """What sampling temperature to use, between 0 and 2.
 
     Higher values like 0.8 will make the output more random, while lower values like
-    0.2 will make it more focused and deterministic.
-
-    We generally recommend altering this or `top_p` but not both.
+    0.2 will make it more focused and deterministic. We generally recommend altering
+    this or `top_p` but not both.
     """
 
     tool_choice: ChatCompletionToolChoiceOptionParam
@@ -218,7 +287,7 @@ class CompletionCreateParamsBase(TypedDict, total=False):
     """
     A unique identifier representing your end-user, which can help OpenAI to monitor
     and detect abuse.
-    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
     """
 
 
@@ -239,7 +308,7 @@ class Function(TypedDict, total=False):
     how to call the function.
     """
 
-    parameters: shared_params.FunctionParameters
+    parameters: FunctionParameters
     """The parameters the functions accepts, described as a JSON Schema object.
 
     See the [guide](https://platform.openai.com/docs/guides/function-calling) for
@@ -251,12 +320,10 @@ class Function(TypedDict, total=False):
     """
 
 
-ResponseFormat: TypeAlias = Union[
-    shared_params.ResponseFormatText, shared_params.ResponseFormatJSONObject, shared_params.ResponseFormatJSONSchema
-]
+ResponseFormat: TypeAlias = Union[ResponseFormatText, ResponseFormatJSONObject, ResponseFormatJSONSchema]
 
 
-class CompletionCreateParamsNonStreaming(CompletionCreateParamsBase):
+class CompletionCreateParamsNonStreaming(CompletionCreateParamsBase, total=False):
     stream: Optional[Literal[False]]
     """If set, partial message deltas will be sent, like in ChatGPT.
 
diff --git a/portkey_ai/_vendor/openai/types/chat_model.py b/portkey_ai/_vendor/openai/types/chat_model.py
index 686f26b7..e1ac4643 100644
--- a/portkey_ai/_vendor/openai/types/chat_model.py
+++ b/portkey_ai/_vendor/openai/types/chat_model.py
@@ -5,9 +5,22 @@
 __all__ = ["ChatModel"]
 
 ChatModel: TypeAlias = Literal[
+    "o1",
+    "o1-2024-12-17",
+    "o1-preview",
+    "o1-preview-2024-09-12",
+    "o1-mini",
+    "o1-mini-2024-09-12",
     "gpt-4o",
+    "gpt-4o-2024-11-20",
     "gpt-4o-2024-08-06",
     "gpt-4o-2024-05-13",
+    "gpt-4o-audio-preview",
+    "gpt-4o-audio-preview-2024-10-01",
+    "gpt-4o-audio-preview-2024-12-17",
+    "gpt-4o-mini-audio-preview",
+    "gpt-4o-mini-audio-preview-2024-12-17",
+    "chatgpt-4o-latest",
     "gpt-4o-mini",
     "gpt-4o-mini-2024-07-18",
     "gpt-4-turbo",
diff --git a/portkey_ai/_vendor/openai/types/completion_create_params.py b/portkey_ai/_vendor/openai/types/completion_create_params.py
index 9fe22fe3..fdb1680d 100644
--- a/portkey_ai/_vendor/openai/types/completion_create_params.py
+++ b/portkey_ai/_vendor/openai/types/completion_create_params.py
@@ -17,8 +17,8 @@ class CompletionCreateParamsBase(TypedDict, total=False):
     You can use the
     [List models](https://platform.openai.com/docs/api-reference/models/list) API to
     see all of your available models, or see our
-    [Model overview](https://platform.openai.com/docs/models/overview) for
-    descriptions of them.
+    [Model overview](https://platform.openai.com/docs/models) for descriptions of
+    them.
     """
 
     prompt: Required[Union[str, List[str], Iterable[int], Iterable[Iterable[int]], None]]
@@ -53,7 +53,7 @@ class CompletionCreateParamsBase(TypedDict, total=False):
     Positive values penalize new tokens based on their existing frequency in the
     text so far, decreasing the model's likelihood to repeat the same line verbatim.
 
-    [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+    [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
     """
 
     logit_bias: Optional[Dict[str, int]]
@@ -106,7 +106,7 @@ class CompletionCreateParamsBase(TypedDict, total=False):
     Positive values penalize new tokens based on whether they appear in the text so
     far, increasing the model's likelihood to talk about new topics.
 
-    [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)
+    [See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation)
     """
 
     seed: Optional[int]
@@ -156,11 +156,11 @@ class CompletionCreateParamsBase(TypedDict, total=False):
     """
     A unique identifier representing your end-user, which can help OpenAI to monitor
     and detect abuse.
-    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
     """
 
 
-class CompletionCreateParamsNonStreaming(CompletionCreateParamsBase):
+class CompletionCreateParamsNonStreaming(CompletionCreateParamsBase, total=False):
     stream: Optional[Literal[False]]
     """Whether to stream back partial progress.
 
diff --git a/portkey_ai/_vendor/openai/types/completion_usage.py b/portkey_ai/_vendor/openai/types/completion_usage.py
index ac09afd4..d8c4e84c 100644
--- a/portkey_ai/_vendor/openai/types/completion_usage.py
+++ b/portkey_ai/_vendor/openai/types/completion_usage.py
@@ -1,9 +1,40 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
+from typing import Optional
 
 from .._models import BaseModel
 
-__all__ = ["CompletionUsage"]
+__all__ = ["CompletionUsage", "CompletionTokensDetails", "PromptTokensDetails"]
+
+
+class CompletionTokensDetails(BaseModel):
+    accepted_prediction_tokens: Optional[int] = None
+    """
+    When using Predicted Outputs, the number of tokens in the prediction that
+    appeared in the completion.
+    """
+
+    audio_tokens: Optional[int] = None
+    """Audio input tokens generated by the model."""
+
+    reasoning_tokens: Optional[int] = None
+    """Tokens generated by the model for reasoning."""
+
+    rejected_prediction_tokens: Optional[int] = None
+    """
+    When using Predicted Outputs, the number of tokens in the prediction that did
+    not appear in the completion. However, like reasoning tokens, these tokens are
+    still counted in the total completion tokens for purposes of billing, output,
+    and context window limits.
+    """
+
+
+class PromptTokensDetails(BaseModel):
+    audio_tokens: Optional[int] = None
+    """Audio input tokens present in the prompt."""
+
+    cached_tokens: Optional[int] = None
+    """Cached tokens present in the prompt."""
 
 
 class CompletionUsage(BaseModel):
@@ -15,3 +46,9 @@ class CompletionUsage(BaseModel):
 
     total_tokens: int
     """Total number of tokens used in the request (prompt + completion)."""
+
+    completion_tokens_details: Optional[CompletionTokensDetails] = None
+    """Breakdown of tokens used in a completion."""
+
+    prompt_tokens_details: Optional[PromptTokensDetails] = None
+    """Breakdown of tokens used in the prompt."""
diff --git a/portkey_ai/_vendor/openai/types/embedding_create_params.py b/portkey_ai/_vendor/openai/types/embedding_create_params.py
index 930b3b79..13857628 100644
--- a/portkey_ai/_vendor/openai/types/embedding_create_params.py
+++ b/portkey_ai/_vendor/openai/types/embedding_create_params.py
@@ -5,6 +5,8 @@
 from typing import List, Union, Iterable
 from typing_extensions import Literal, Required, TypedDict
 
+from .embedding_model import EmbeddingModel
+
 __all__ = ["EmbeddingCreateParams"]
 
 
@@ -20,14 +22,14 @@ class EmbeddingCreateParams(TypedDict, total=False):
     for counting tokens.
     """
 
-    model: Required[Union[str, Literal["text-embedding-ada-002", "text-embedding-3-small", "text-embedding-3-large"]]]
+    model: Required[Union[str, EmbeddingModel]]
     """ID of the model to use.
 
     You can use the
     [List models](https://platform.openai.com/docs/api-reference/models/list) API to
     see all of your available models, or see our
-    [Model overview](https://platform.openai.com/docs/models/overview) for
-    descriptions of them.
+    [Model overview](https://platform.openai.com/docs/models) for descriptions of
+    them.
     """
 
     dimensions: int
@@ -46,5 +48,5 @@ class EmbeddingCreateParams(TypedDict, total=False):
     """
     A unique identifier representing your end-user, which can help OpenAI to monitor
     and detect abuse.
-    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
     """
diff --git a/portkey_ai/_vendor/openai/types/embedding_model.py b/portkey_ai/_vendor/openai/types/embedding_model.py
new file mode 100644
index 00000000..075ff976
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/embedding_model.py
@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal, TypeAlias
+
+__all__ = ["EmbeddingModel"]
+
+EmbeddingModel: TypeAlias = Literal["text-embedding-ada-002", "text-embedding-3-small", "text-embedding-3-large"]
diff --git a/portkey_ai/_vendor/openai/types/file_create_params.py b/portkey_ai/_vendor/openai/types/file_create_params.py
index 8b1c296f..ecf75033 100644
--- a/portkey_ai/_vendor/openai/types/file_create_params.py
+++ b/portkey_ai/_vendor/openai/types/file_create_params.py
@@ -2,9 +2,10 @@
 
 from __future__ import annotations
 
-from typing_extensions import Literal, Required, TypedDict
+from typing_extensions import Required, TypedDict
 
 from .._types import FileTypes
+from .file_purpose import FilePurpose
 
 __all__ = ["FileCreateParams"]
 
@@ -13,7 +14,7 @@ class FileCreateParams(TypedDict, total=False):
     file: Required[FileTypes]
     """The File object (not file name) to be uploaded."""
 
-    purpose: Required[Literal["assistants", "batch", "fine-tune", "vision"]]
+    purpose: Required[FilePurpose]
     """The intended purpose of the uploaded file.
 
     Use "assistants" for
diff --git a/portkey_ai/_vendor/openai/types/file_list_params.py b/portkey_ai/_vendor/openai/types/file_list_params.py
index 212eca13..058d874c 100644
--- a/portkey_ai/_vendor/openai/types/file_list_params.py
+++ b/portkey_ai/_vendor/openai/types/file_list_params.py
@@ -2,11 +2,32 @@
 
 from __future__ import annotations
 
-from typing_extensions import TypedDict
+from typing_extensions import Literal, TypedDict
 
 __all__ = ["FileListParams"]
 
 
 class FileListParams(TypedDict, total=False):
+    after: str
+    """A cursor for use in pagination.
+
+    `after` is an object ID that defines your place in the list. For instance, if
+    you make a list request and receive 100 objects, ending with obj_foo, your
+    subsequent call can include after=obj_foo in order to fetch the next page of the
+    list.
+    """
+
+    limit: int
+    """A limit on the number of objects to be returned.
+
+    Limit can range between 1 and 10,000, and the default is 10,000.
+    """
+
+    order: Literal["asc", "desc"]
+    """Sort order by the `created_at` timestamp of the objects.
+
+    `asc` for ascending order and `desc` for descending order.
+    """
+
     purpose: str
     """Only return files with the given purpose."""
diff --git a/portkey_ai/_vendor/openai/types/file_purpose.py b/portkey_ai/_vendor/openai/types/file_purpose.py
new file mode 100644
index 00000000..32dc352c
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/file_purpose.py
@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal, TypeAlias
+
+__all__ = ["FilePurpose"]
+
+FilePurpose: TypeAlias = Literal["assistants", "batch", "fine-tune", "vision"]
diff --git a/portkey_ai/_vendor/openai/types/fine_tuning/fine_tuning_job.py b/portkey_ai/_vendor/openai/types/fine_tuning/fine_tuning_job.py
index 7ac87927..f5a11c21 100644
--- a/portkey_ai/_vendor/openai/types/fine_tuning/fine_tuning_job.py
+++ b/portkey_ai/_vendor/openai/types/fine_tuning/fine_tuning_job.py
@@ -6,7 +6,16 @@
 from ..._models import BaseModel
 from .fine_tuning_job_wandb_integration_object import FineTuningJobWandbIntegrationObject
 
-__all__ = ["FineTuningJob", "Error", "Hyperparameters"]
+__all__ = [
+    "FineTuningJob",
+    "Error",
+    "Hyperparameters",
+    "Method",
+    "MethodDpo",
+    "MethodDpoHyperparameters",
+    "MethodSupervised",
+    "MethodSupervisedHyperparameters",
+]
 
 
 class Error(BaseModel):
@@ -24,15 +33,96 @@ class Error(BaseModel):
 
 
 class Hyperparameters(BaseModel):
-    n_epochs: Union[Literal["auto"], int]
+    batch_size: Union[Literal["auto"], int, None] = None
+    """Number of examples in each batch.
+
+    A larger batch size means that model parameters are updated less frequently, but
+    with lower variance.
+    """
+
+    learning_rate_multiplier: Union[Literal["auto"], float, None] = None
+    """Scaling factor for the learning rate.
+
+    A smaller learning rate may be useful to avoid overfitting.
+    """
+
+    n_epochs: Union[Literal["auto"], int, None] = None
+    """The number of epochs to train the model for.
+
+    An epoch refers to one full cycle through the training dataset.
+    """
+
+
+class MethodDpoHyperparameters(BaseModel):
+    batch_size: Union[Literal["auto"], int, None] = None
+    """Number of examples in each batch.
+
+    A larger batch size means that model parameters are updated less frequently, but
+    with lower variance.
+    """
+
+    beta: Union[Literal["auto"], float, None] = None
+    """The beta value for the DPO method.
+
+    A higher beta value will increase the weight of the penalty between the policy
+    and reference model.
+    """
+
+    learning_rate_multiplier: Union[Literal["auto"], float, None] = None
+    """Scaling factor for the learning rate.
+
+    A smaller learning rate may be useful to avoid overfitting.
+    """
+
+    n_epochs: Union[Literal["auto"], int, None] = None
     """The number of epochs to train the model for.
 
-    An epoch refers to one full cycle through the training dataset. "auto" decides
-    the optimal number of epochs based on the size of the dataset. If setting the
-    number manually, we support any number between 1 and 50 epochs.
+    An epoch refers to one full cycle through the training dataset.
     """
 
 
+class MethodDpo(BaseModel):
+    hyperparameters: Optional[MethodDpoHyperparameters] = None
+    """The hyperparameters used for the fine-tuning job."""
+
+
+class MethodSupervisedHyperparameters(BaseModel):
+    batch_size: Union[Literal["auto"], int, None] = None
+    """Number of examples in each batch.
+
+    A larger batch size means that model parameters are updated less frequently, but
+    with lower variance.
+    """
+
+    learning_rate_multiplier: Union[Literal["auto"], float, None] = None
+    """Scaling factor for the learning rate.
+
+    A smaller learning rate may be useful to avoid overfitting.
+    """
+
+    n_epochs: Union[Literal["auto"], int, None] = None
+    """The number of epochs to train the model for.
+
+    An epoch refers to one full cycle through the training dataset.
+    """
+
+
+class MethodSupervised(BaseModel):
+    hyperparameters: Optional[MethodSupervisedHyperparameters] = None
+    """The hyperparameters used for the fine-tuning job."""
+
+
+class Method(BaseModel):
+    dpo: Optional[MethodDpo] = None
+    """Configuration for the DPO fine-tuning method."""
+
+    supervised: Optional[MethodSupervised] = None
+    """Configuration for the supervised fine-tuning method."""
+
+    type: Optional[Literal["supervised", "dpo"]] = None
+    """The type of method. Is either `supervised` or `dpo`."""
+
+
 class FineTuningJob(BaseModel):
     id: str
     """The object identifier, which can be referenced in the API endpoints."""
@@ -61,8 +151,7 @@ class FineTuningJob(BaseModel):
     hyperparameters: Hyperparameters
     """The hyperparameters used for the fine-tuning job.
 
-    See the [fine-tuning guide](https://platform.openai.com/docs/guides/fine-tuning)
-    for more details.
+    This value will only be returned when running `supervised` jobs.
     """
 
     model: str
@@ -118,3 +207,6 @@ class FineTuningJob(BaseModel):
 
     integrations: Optional[List[FineTuningJobWandbIntegrationObject]] = None
     """A list of integrations to enable for this fine-tuning job."""
+
+    method: Optional[Method] = None
+    """The method used for fine-tuning."""
diff --git a/portkey_ai/_vendor/openai/types/fine_tuning/fine_tuning_job_event.py b/portkey_ai/_vendor/openai/types/fine_tuning/fine_tuning_job_event.py
index 2d204bb9..1d728bd7 100644
--- a/portkey_ai/_vendor/openai/types/fine_tuning/fine_tuning_job_event.py
+++ b/portkey_ai/_vendor/openai/types/fine_tuning/fine_tuning_job_event.py
@@ -1,5 +1,7 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
+import builtins
+from typing import Optional
 from typing_extensions import Literal
 
 from ..._models import BaseModel
@@ -9,11 +11,22 @@
 
 class FineTuningJobEvent(BaseModel):
     id: str
+    """The object identifier."""
 
     created_at: int
+    """The Unix timestamp (in seconds) for when the fine-tuning job was created."""
 
     level: Literal["info", "warn", "error"]
+    """The log level of the event."""
 
     message: str
+    """The message of the event."""
 
     object: Literal["fine_tuning.job.event"]
+    """The object type, which is always "fine_tuning.job.event"."""
+
+    data: Optional[builtins.object] = None
+    """The data associated with the event."""
+
+    type: Optional[Literal["message", "metrics"]] = None
+    """The type of event."""
diff --git a/portkey_ai/_vendor/openai/types/fine_tuning/job_create_params.py b/portkey_ai/_vendor/openai/types/fine_tuning/job_create_params.py
index e9be2ef1..09c3f857 100644
--- a/portkey_ai/_vendor/openai/types/fine_tuning/job_create_params.py
+++ b/portkey_ai/_vendor/openai/types/fine_tuning/job_create_params.py
@@ -5,7 +5,17 @@
 from typing import List, Union, Iterable, Optional
 from typing_extensions import Literal, Required, TypedDict
 
-__all__ = ["JobCreateParams", "Hyperparameters", "Integration", "IntegrationWandb"]
+__all__ = [
+    "JobCreateParams",
+    "Hyperparameters",
+    "Integration",
+    "IntegrationWandb",
+    "Method",
+    "MethodDpo",
+    "MethodDpoHyperparameters",
+    "MethodSupervised",
+    "MethodSupervisedHyperparameters",
+]
 
 
 class JobCreateParams(TypedDict, total=False):
@@ -13,7 +23,7 @@ class JobCreateParams(TypedDict, total=False):
     """The name of the model to fine-tune.
 
     You can select one of the
-    [supported models](https://platform.openai.com/docs/guides/fine-tuning/which-models-can-be-fine-tuned).
+    [supported models](https://platform.openai.com/docs/guides/fine-tuning#which-models-can-be-fine-tuned).
     """
 
     training_file: Required[str]
@@ -26,8 +36,10 @@ class JobCreateParams(TypedDict, total=False):
     your file with the purpose `fine-tune`.
 
     The contents of the file should differ depending on if the model uses the
-    [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input) or
+    [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input),
     [completions](https://platform.openai.com/docs/api-reference/fine-tuning/completions-input)
+    format, or if the fine-tuning method uses the
+    [preference](https://platform.openai.com/docs/api-reference/fine-tuning/preference-input)
     format.
 
     See the [fine-tuning guide](https://platform.openai.com/docs/guides/fine-tuning)
@@ -35,11 +47,17 @@ class JobCreateParams(TypedDict, total=False):
     """
 
     hyperparameters: Hyperparameters
-    """The hyperparameters used for the fine-tuning job."""
+    """
+    The hyperparameters used for the fine-tuning job. This value is now deprecated
+    in favor of `method`, and should be passed in under the `method` parameter.
+    """
 
     integrations: Optional[Iterable[Integration]]
     """A list of integrations to enable for your fine-tuning job."""
 
+    method: Method
+    """The method used for fine-tuning."""
+
     seed: Optional[int]
     """The seed controls the reproducibility of the job.
 
@@ -50,7 +68,7 @@ class JobCreateParams(TypedDict, total=False):
 
     suffix: Optional[str]
     """
-    A string of up to 18 characters that will be added to your fine-tuned model
+    A string of up to 64 characters that will be added to your fine-tuned model
     name.
 
     For example, a `suffix` of "custom-model-name" would produce a model name like
@@ -134,3 +152,73 @@ class Integration(TypedDict, total=False):
     can set an explicit display name for your run, add tags to your run, and set a
     default entity (team, username, etc) to be associated with your run.
     """
+
+
+class MethodDpoHyperparameters(TypedDict, total=False):
+    batch_size: Union[Literal["auto"], int]
+    """Number of examples in each batch.
+
+    A larger batch size means that model parameters are updated less frequently, but
+    with lower variance.
+    """
+
+    beta: Union[Literal["auto"], float]
+    """The beta value for the DPO method.
+
+    A higher beta value will increase the weight of the penalty between the policy
+    and reference model.
+    """
+
+    learning_rate_multiplier: Union[Literal["auto"], float]
+    """Scaling factor for the learning rate.
+
+    A smaller learning rate may be useful to avoid overfitting.
+    """
+
+    n_epochs: Union[Literal["auto"], int]
+    """The number of epochs to train the model for.
+
+    An epoch refers to one full cycle through the training dataset.
+    """
+
+
+class MethodDpo(TypedDict, total=False):
+    hyperparameters: MethodDpoHyperparameters
+    """The hyperparameters used for the fine-tuning job."""
+
+
+class MethodSupervisedHyperparameters(TypedDict, total=False):
+    batch_size: Union[Literal["auto"], int]
+    """Number of examples in each batch.
+
+    A larger batch size means that model parameters are updated less frequently, but
+    with lower variance.
+    """
+
+    learning_rate_multiplier: Union[Literal["auto"], float]
+    """Scaling factor for the learning rate.
+
+    A smaller learning rate may be useful to avoid overfitting.
+    """
+
+    n_epochs: Union[Literal["auto"], int]
+    """The number of epochs to train the model for.
+
+    An epoch refers to one full cycle through the training dataset.
+    """
+
+
+class MethodSupervised(TypedDict, total=False):
+    hyperparameters: MethodSupervisedHyperparameters
+    """The hyperparameters used for the fine-tuning job."""
+
+
+class Method(TypedDict, total=False):
+    dpo: MethodDpo
+    """Configuration for the DPO fine-tuning method."""
+
+    supervised: MethodSupervised
+    """Configuration for the supervised fine-tuning method."""
+
+    type: Literal["supervised", "dpo"]
+    """The type of method. Is either `supervised` or `dpo`."""
diff --git a/portkey_ai/_vendor/openai/types/image_create_variation_params.py b/portkey_ai/_vendor/openai/types/image_create_variation_params.py
index d6ecf0f1..d20f6729 100644
--- a/portkey_ai/_vendor/openai/types/image_create_variation_params.py
+++ b/portkey_ai/_vendor/openai/types/image_create_variation_params.py
@@ -47,5 +47,5 @@ class ImageCreateVariationParams(TypedDict, total=False):
     """
     A unique identifier representing your end-user, which can help OpenAI to monitor
     and detect abuse.
-    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
     """
diff --git a/portkey_ai/_vendor/openai/types/image_edit_params.py b/portkey_ai/_vendor/openai/types/image_edit_params.py
index a596a869..1cb10611 100644
--- a/portkey_ai/_vendor/openai/types/image_edit_params.py
+++ b/portkey_ai/_vendor/openai/types/image_edit_params.py
@@ -58,5 +58,5 @@ class ImageEditParams(TypedDict, total=False):
     """
     A unique identifier representing your end-user, which can help OpenAI to monitor
     and detect abuse.
-    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
     """
diff --git a/portkey_ai/_vendor/openai/types/image_generate_params.py b/portkey_ai/_vendor/openai/types/image_generate_params.py
index 307adeb3..c88c45f5 100644
--- a/portkey_ai/_vendor/openai/types/image_generate_params.py
+++ b/portkey_ai/_vendor/openai/types/image_generate_params.py
@@ -61,5 +61,5 @@ class ImageGenerateParams(TypedDict, total=False):
     """
     A unique identifier representing your end-user, which can help OpenAI to monitor
     and detect abuse.
-    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+    [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
     """
diff --git a/portkey_ai/_vendor/openai/types/moderation.py b/portkey_ai/_vendor/openai/types/moderation.py
index 5aa69182..e4ec182c 100644
--- a/portkey_ai/_vendor/openai/types/moderation.py
+++ b/portkey_ai/_vendor/openai/types/moderation.py
@@ -1,11 +1,13 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
+from typing import List
+from typing_extensions import Literal
 
 from pydantic import Field as FieldInfo
 
 from .._models import BaseModel
 
-__all__ = ["Moderation", "Categories", "CategoryScores"]
+__all__ = ["Moderation", "Categories", "CategoryAppliedInputTypes", "CategoryScores"]
 
 
 class Categories(BaseModel):
@@ -36,6 +38,20 @@ class Categories(BaseModel):
     orientation, disability status, or caste.
     """
 
+    illicit: bool
+    """
+    Content that includes instructions or advice that facilitate the planning or
+    execution of wrongdoing, or that gives advice or instruction on how to commit
+    illicit acts. For example, "how to shoplift" would fit this category.
+    """
+
+    illicit_violent: bool = FieldInfo(alias="illicit/violent")
+    """
+    Content that includes instructions or advice that facilitate the planning or
+    execution of wrongdoing that also includes violence, or that gives advice or
+    instruction on the procurement of any weapon.
+    """
+
     self_harm: bool = FieldInfo(alias="self-harm")
     """
     Content that promotes, encourages, or depicts acts of self-harm, such as
@@ -72,6 +88,47 @@ class Categories(BaseModel):
     """Content that depicts death, violence, or physical injury in graphic detail."""
 
 
+class CategoryAppliedInputTypes(BaseModel):
+    harassment: List[Literal["text"]]
+    """The applied input type(s) for the category 'harassment'."""
+
+    harassment_threatening: List[Literal["text"]] = FieldInfo(alias="harassment/threatening")
+    """The applied input type(s) for the category 'harassment/threatening'."""
+
+    hate: List[Literal["text"]]
+    """The applied input type(s) for the category 'hate'."""
+
+    hate_threatening: List[Literal["text"]] = FieldInfo(alias="hate/threatening")
+    """The applied input type(s) for the category 'hate/threatening'."""
+
+    illicit: List[Literal["text"]]
+    """The applied input type(s) for the category 'illicit'."""
+
+    illicit_violent: List[Literal["text"]] = FieldInfo(alias="illicit/violent")
+    """The applied input type(s) for the category 'illicit/violent'."""
+
+    self_harm: List[Literal["text", "image"]] = FieldInfo(alias="self-harm")
+    """The applied input type(s) for the category 'self-harm'."""
+
+    self_harm_instructions: List[Literal["text", "image"]] = FieldInfo(alias="self-harm/instructions")
+    """The applied input type(s) for the category 'self-harm/instructions'."""
+
+    self_harm_intent: List[Literal["text", "image"]] = FieldInfo(alias="self-harm/intent")
+    """The applied input type(s) for the category 'self-harm/intent'."""
+
+    sexual: List[Literal["text", "image"]]
+    """The applied input type(s) for the category 'sexual'."""
+
+    sexual_minors: List[Literal["text"]] = FieldInfo(alias="sexual/minors")
+    """The applied input type(s) for the category 'sexual/minors'."""
+
+    violence: List[Literal["text", "image"]]
+    """The applied input type(s) for the category 'violence'."""
+
+    violence_graphic: List[Literal["text", "image"]] = FieldInfo(alias="violence/graphic")
+    """The applied input type(s) for the category 'violence/graphic'."""
+
+
 class CategoryScores(BaseModel):
     harassment: float
     """The score for the category 'harassment'."""
@@ -85,6 +142,12 @@ class CategoryScores(BaseModel):
     hate_threatening: float = FieldInfo(alias="hate/threatening")
     """The score for the category 'hate/threatening'."""
 
+    illicit: float
+    """The score for the category 'illicit'."""
+
+    illicit_violent: float = FieldInfo(alias="illicit/violent")
+    """The score for the category 'illicit/violent'."""
+
     self_harm: float = FieldInfo(alias="self-harm")
     """The score for the category 'self-harm'."""
 
@@ -111,6 +174,11 @@ class Moderation(BaseModel):
     categories: Categories
     """A list of the categories, and whether they are flagged or not."""
 
+    category_applied_input_types: CategoryAppliedInputTypes
+    """
+    A list of the categories along with the input type(s) that the score applies to.
+    """
+
     category_scores: CategoryScores
     """A list of the categories along with their scores as predicted by model."""
 
diff --git a/portkey_ai/_vendor/openai/types/moderation_create_params.py b/portkey_ai/_vendor/openai/types/moderation_create_params.py
index 33768219..3ea2f3cd 100644
--- a/portkey_ai/_vendor/openai/types/moderation_create_params.py
+++ b/portkey_ai/_vendor/openai/types/moderation_create_params.py
@@ -2,26 +2,28 @@
 
 from __future__ import annotations
 
-from typing import List, Union
+from typing import List, Union, Iterable
 from typing_extensions import Required, TypedDict
 
 from .moderation_model import ModerationModel
+from .moderation_multi_modal_input_param import ModerationMultiModalInputParam
 
 __all__ = ["ModerationCreateParams"]
 
 
 class ModerationCreateParams(TypedDict, total=False):
-    input: Required[Union[str, List[str]]]
-    """The input text to classify"""
+    input: Required[Union[str, List[str], Iterable[ModerationMultiModalInputParam]]]
+    """Input (or inputs) to classify.
 
-    model: Union[str, ModerationModel]
+    Can be a single string, an array of strings, or an array of multi-modal input
+    objects similar to other models.
     """
-    Two content moderations models are available: `text-moderation-stable` and
-    `text-moderation-latest`.
-
-    The default is `text-moderation-latest` which will be automatically upgraded
-    over time. This ensures you are always using our most accurate model. If you use
-    `text-moderation-stable`, we will provide advanced notice before updating the
-    model. Accuracy of `text-moderation-stable` may be slightly lower than for
-    `text-moderation-latest`.
+
+    model: Union[str, ModerationModel]
+    """The content moderation model you would like to use.
+
+    Learn more in
+    [the moderation guide](https://platform.openai.com/docs/guides/moderation), and
+    learn about available models
+    [here](https://platform.openai.com/docs/models#moderation).
     """
diff --git a/portkey_ai/_vendor/openai/types/moderation_image_url_input_param.py b/portkey_ai/_vendor/openai/types/moderation_image_url_input_param.py
new file mode 100644
index 00000000..9a69a6a2
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/moderation_image_url_input_param.py
@@ -0,0 +1,20 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["ModerationImageURLInputParam", "ImageURL"]
+
+
+class ImageURL(TypedDict, total=False):
+    url: Required[str]
+    """Either a URL of the image or the base64 encoded image data."""
+
+
+class ModerationImageURLInputParam(TypedDict, total=False):
+    image_url: Required[ImageURL]
+    """Contains either an image URL or a data URL for a base64 encoded image."""
+
+    type: Required[Literal["image_url"]]
+    """Always `image_url`."""
diff --git a/portkey_ai/_vendor/openai/types/moderation_model.py b/portkey_ai/_vendor/openai/types/moderation_model.py
index f549aeeb..64954c45 100644
--- a/portkey_ai/_vendor/openai/types/moderation_model.py
+++ b/portkey_ai/_vendor/openai/types/moderation_model.py
@@ -4,4 +4,6 @@
 
 __all__ = ["ModerationModel"]
 
-ModerationModel: TypeAlias = Literal["text-moderation-latest", "text-moderation-stable"]
+ModerationModel: TypeAlias = Literal[
+    "omni-moderation-latest", "omni-moderation-2024-09-26", "text-moderation-latest", "text-moderation-stable"
+]
diff --git a/portkey_ai/_vendor/openai/types/moderation_multi_modal_input_param.py b/portkey_ai/_vendor/openai/types/moderation_multi_modal_input_param.py
new file mode 100644
index 00000000..4314e7b0
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/moderation_multi_modal_input_param.py
@@ -0,0 +1,13 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union
+from typing_extensions import TypeAlias
+
+from .moderation_text_input_param import ModerationTextInputParam
+from .moderation_image_url_input_param import ModerationImageURLInputParam
+
+__all__ = ["ModerationMultiModalInputParam"]
+
+ModerationMultiModalInputParam: TypeAlias = Union[ModerationImageURLInputParam, ModerationTextInputParam]
diff --git a/portkey_ai/_vendor/openai/types/moderation_text_input_param.py b/portkey_ai/_vendor/openai/types/moderation_text_input_param.py
new file mode 100644
index 00000000..e5da5333
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/moderation_text_input_param.py
@@ -0,0 +1,15 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["ModerationTextInputParam"]
+
+
+class ModerationTextInputParam(TypedDict, total=False):
+    text: Required[str]
+    """A string of text to classify."""
+
+    type: Required[Literal["text"]]
+    """Always `text`."""
diff --git a/portkey_ai/_vendor/openai/types/shared_params/function_definition.py b/portkey_ai/_vendor/openai/types/shared_params/function_definition.py
index f41392f1..d45ec13f 100644
--- a/portkey_ai/_vendor/openai/types/shared_params/function_definition.py
+++ b/portkey_ai/_vendor/openai/types/shared_params/function_definition.py
@@ -5,7 +5,7 @@
 from typing import Optional
 from typing_extensions import Required, TypedDict
 
-from ...types import shared_params
+from .function_parameters import FunctionParameters
 
 __all__ = ["FunctionDefinition"]
 
@@ -24,7 +24,7 @@ class FunctionDefinition(TypedDict, total=False):
     how to call the function.
     """
 
-    parameters: shared_params.FunctionParameters
+    parameters: FunctionParameters
     """The parameters the functions accepts, described as a JSON Schema object.
 
     See the [guide](https://platform.openai.com/docs/guides/function-calling) for
diff --git a/portkey_ai/_vendor/openai/types/upload_create_params.py b/portkey_ai/_vendor/openai/types/upload_create_params.py
index 3165ebcc..2ebabe6c 100644
--- a/portkey_ai/_vendor/openai/types/upload_create_params.py
+++ b/portkey_ai/_vendor/openai/types/upload_create_params.py
@@ -2,7 +2,9 @@
 
 from __future__ import annotations
 
-from typing_extensions import Literal, Required, TypedDict
+from typing_extensions import Required, TypedDict
+
+from .file_purpose import FilePurpose
 
 __all__ = ["UploadCreateParams"]
 
@@ -21,7 +23,7 @@ class UploadCreateParams(TypedDict, total=False):
     supported MIME types for assistants and vision.
     """
 
-    purpose: Required[Literal["assistants", "batch", "fine-tune", "vision"]]
+    purpose: Required[FilePurpose]
     """The intended purpose of the uploaded file.
 
     See the
diff --git a/portkey_ai/_vendor/openai/types/websocket_connection_options.py b/portkey_ai/_vendor/openai/types/websocket_connection_options.py
new file mode 100644
index 00000000..40fd24ab
--- /dev/null
+++ b/portkey_ai/_vendor/openai/types/websocket_connection_options.py
@@ -0,0 +1,36 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+from typing_extensions import Sequence, TypedDict
+
+if TYPE_CHECKING:
+    from websockets import Subprotocol
+    from websockets.extensions import ClientExtensionFactory
+
+
+class WebsocketConnectionOptions(TypedDict, total=False):
+    """Websocket connection options copied from `websockets`.
+
+    For example: https://websockets.readthedocs.io/en/stable/reference/asyncio/client.html#websockets.asyncio.client.connect
+    """
+
+    extensions: Sequence[ClientExtensionFactory] | None
+    """List of supported extensions, in order in which they should be negotiated and run."""
+
+    subprotocols: Sequence[Subprotocol] | None
+    """List of supported subprotocols, in order of decreasing preference."""
+
+    compression: str | None
+    """The “permessage-deflate” extension is enabled by default. Set compression to None to disable it. See the [compression guide](https://websockets.readthedocs.io/en/stable/topics/compression.html) for details."""
+
+    # limits
+    max_size: int | None
+    """Maximum size of incoming messages in bytes. None disables the limit."""
+
+    max_queue: int | None | tuple[int | None, int | None]
+    """High-water mark of the buffer where frames are received. It defaults to 16 frames. The low-water mark defaults to max_queue // 4. You may pass a (high, low) tuple to set the high-water and low-water marks. If you want to disable flow control entirely, you may set it to None, although that’s a bad idea."""
+
+    write_limit: int | tuple[int, int | None]
+    """High-water mark of write buffer in bytes. It is passed to set_write_buffer_limits(). It defaults to 32 KiB. You may pass a (high, low) tuple to set the high-water and low-water marks."""
diff --git a/portkey_ai/api_resources/__init__.py b/portkey_ai/api_resources/__init__.py
index ebacc055..715ff4c4 100644
--- a/portkey_ai/api_resources/__init__.py
+++ b/portkey_ai/api_resources/__init__.py
@@ -77,6 +77,10 @@
     AsyncVirtualKeys,
     Logs,
     AsyncLogs,
+    BetaRealtime,
+    AsyncBetaRealtime,
+    BetaSessions,
+    AsyncBetaSessions,
 )
 from .utils import (
     Modes,
@@ -189,4 +193,8 @@
     "AsyncVirtualKeys",
     "Logs",
     "AsyncLogs",
+    "BetaRealtime",
+    "AsyncBetaRealtime",
+    "BetaSessions",
+    "AsyncBetaSessions",
 ]
diff --git a/portkey_ai/api_resources/apis/__init__.py b/portkey_ai/api_resources/apis/__init__.py
index 425c3300..599f413b 100644
--- a/portkey_ai/api_resources/apis/__init__.py
+++ b/portkey_ai/api_resources/apis/__init__.py
@@ -74,6 +74,13 @@
     AsyncBetaCompletions,
 )
 
+from .beta_realtime import (
+    BetaRealtime,
+    AsyncBetaRealtime,
+    BetaSessions,
+    AsyncBetaSessions,
+)
+
 from .uploads import (
     Uploads,
     Parts,
@@ -171,4 +178,8 @@
     "AsyncVirtualKeys",
     "Logs",
     "AsyncLogs",
+    "BetaRealtime",
+    "AsyncBetaRealtime",
+    "BetaSessions",
+    "AsyncBetaSessions",
 ]
diff --git a/portkey_ai/api_resources/apis/audio.py b/portkey_ai/api_resources/apis/audio.py
index b82f7ef8..bfce244a 100644
--- a/portkey_ai/api_resources/apis/audio.py
+++ b/portkey_ai/api_resources/apis/audio.py
@@ -5,7 +5,12 @@
 from portkey_ai.api_resources.client import AsyncPortkey, Portkey
 import typing
 
-from portkey_ai.api_resources.types.audio_types import Transcription, Translation
+from portkey_ai.api_resources.types.audio_types import (
+    Transcription,
+    TranscriptionVerbose,
+    Translation,
+    TranslationVerbose,
+)
 
 
 class Audio(APIResource):
@@ -34,7 +39,7 @@ def create(
         temperature: Union[float, NotGiven] = NOT_GIVEN,
         timestamp_granularities: Union[List[str], NotGiven] = NOT_GIVEN,
         **kwargs
-    ) -> Transcription:
+    ) -> Union[Transcription, TranscriptionVerbose, str]:
         extra_headers = kwargs.pop("extra_headers", {})
         response = self.openai_client.with_raw_response.audio.transcriptions.create(
             file=file,
@@ -47,8 +52,16 @@ def create(
             extra_headers=extra_headers,
             extra_body=kwargs,
         )
-        data = Transcription(**json.loads(response.text))
-        data._headers = response.headers
+
+        if response_format == "verbose_json":
+            data = TranscriptionVerbose(**json.loads(response.text))
+            data._headers = response.headers
+        elif response_format == "json":
+            data = Transcription(**json.loads(response.text))
+            data._headers = response.headers
+        else:
+            data = Transcription(**json.loads(response.text))
+            data._headers = response.headers
 
         return data
 
@@ -64,16 +77,14 @@ def create(
         file: FileTypes,
         model: str,
         prompt: Union[str, NotGiven] = NOT_GIVEN,
-        response_format: Union[str, NotGiven] = NOT_GIVEN,
         temperature: Union[float, NotGiven] = NOT_GIVEN,
         **kwargs
-    ) -> Translation:
+    ) -> Union[Translation, TranslationVerbose, str]:
         extra_headers = kwargs.pop("extra_headers", {})
         response = self.openai_client.with_raw_response.audio.translations.create(
             file=file,
             model=model,
             prompt=prompt,
-            response_format=response_format,
             temperature=temperature,
             extra_headers=extra_headers,
             extra_body=kwargs,
@@ -143,7 +154,7 @@ async def create(
         temperature: Union[float, NotGiven] = NOT_GIVEN,
         timestamp_granularities: Union[List[str], NotGiven] = NOT_GIVEN,
         **kwargs
-    ) -> Transcription:
+    ) -> Union[Transcription, TranscriptionVerbose, str]:
         extra_headers = kwargs.pop("extra_headers", {})
         response = (
             await self.openai_client.with_raw_response.audio.transcriptions.create(
@@ -158,8 +169,16 @@ async def create(
                 extra_body=kwargs,
             )
         )
-        data = Transcription(**json.loads(response.text))
-        data._headers = response.headers
+
+        if response_format == "verbose_json":
+            data = TranscriptionVerbose(**json.loads(response.text))
+            data._headers = response.headers
+        elif response_format == "json":
+            data = Transcription(**json.loads(response.text))
+            data._headers = response.headers
+        else:
+            data = Transcription(**json.loads(response.text))
+            data._headers = response.headers
 
         return data
 
@@ -175,16 +194,14 @@ async def create(
         file: FileTypes,
         model: str,
         prompt: Union[str, NotGiven] = NOT_GIVEN,
-        response_format: Union[str, NotGiven] = NOT_GIVEN,
         temperature: Union[float, NotGiven] = NOT_GIVEN,
         **kwargs
-    ) -> Translation:
+    ) -> Union[Translation, TranslationVerbose, str]:
         extra_headers = kwargs.pop("extra_headers", {})
         response = await self.openai_client.with_raw_response.audio.translations.create(
             file=file,
             model=model,
             prompt=prompt,
-            response_format=response_format,
             temperature=temperature,
             extra_headers=extra_headers,
             extra_body=kwargs,
diff --git a/portkey_ai/api_resources/apis/beta_chat.py b/portkey_ai/api_resources/apis/beta_chat.py
index 4f42cf27..45bf9f54 100644
--- a/portkey_ai/api_resources/apis/beta_chat.py
+++ b/portkey_ai/api_resources/apis/beta_chat.py
@@ -1,4 +1,4 @@
-from typing import Any, Union
+from typing import Any, Dict, List, Union
 from portkey_ai.api_resources.apis.api_resource import APIResource, AsyncAPIResource
 from portkey_ai.api_resources.client import AsyncPortkey, Portkey
 from portkey_ai.api_resources.types.beta_chat_type import ParsedChatCompletion
@@ -24,6 +24,13 @@ def parse(
         model: Union[str, Any] = "portkey-default",
         response_format: Union[Any, NotGiven] = NOT_GIVEN,
         tools: Union[Any, NotGiven] = NOT_GIVEN,
+        audio: Union[Any, NotGiven] = NOT_GIVEN,
+        max_completion_tokens: Union[int, NotGiven] = NOT_GIVEN,
+        metadata: Union[Dict[str, str], NotGiven] = NOT_GIVEN,
+        modalities: Union[List[Any], NotGiven] = NOT_GIVEN,
+        prediction: Union[Any, NotGiven] = NOT_GIVEN,
+        reasoning_effort: Union[Any, NotGiven] = NOT_GIVEN,
+        store: Union[bool, NotGiven] = NOT_GIVEN,
         **kwargs: Any,
     ) -> ParsedChatCompletion:
         response = self.openai_client.beta.chat.completions.parse(
@@ -31,6 +38,13 @@ def parse(
             model=model,
             response_format=response_format,
             tools=tools,
+            audio=audio,
+            max_completion_tokens=max_completion_tokens,
+            metadata=metadata,
+            modalities=modalities,
+            prediction=prediction,
+            reasoning_effort=reasoning_effort,
+            store=store,
             extra_body=kwargs,
         )
         return response  # type: ignore [return-value]
@@ -42,6 +56,13 @@ def stream(
         model: Union[str, Any] = "portkey-default",
         response_format: Union[Any, NotGiven] = NOT_GIVEN,
         tools: Union[Any, NotGiven] = NOT_GIVEN,
+        audio: Union[Any, NotGiven] = NOT_GIVEN,
+        max_completion_tokens: Union[int, NotGiven] = NOT_GIVEN,
+        metadata: Union[Dict[str, str], NotGiven] = NOT_GIVEN,
+        modalities: Union[List[Any], NotGiven] = NOT_GIVEN,
+        prediction: Union[Any, NotGiven] = NOT_GIVEN,
+        reasoning_effort: Union[Any, NotGiven] = NOT_GIVEN,
+        store: Union[bool, NotGiven] = NOT_GIVEN,
         **kwargs: Any,
     ) -> Any:
         with self.openai_client.beta.chat.completions.stream(
@@ -49,6 +70,13 @@ def stream(
             model=model,
             response_format=response_format,
             tools=tools,
+            audio=audio,
+            max_completion_tokens=max_completion_tokens,
+            metadata=metadata,
+            modalities=modalities,
+            prediction=prediction,
+            reasoning_effort=reasoning_effort,
+            store=store,
             extra_body=kwargs,
         ) as stream:
             for event in stream:
@@ -82,6 +110,13 @@ async def parse(
         model: Union[str, Any] = "portkey-default",
         response_format: Union[Any, NotGiven] = NOT_GIVEN,
         tools: Union[Any, NotGiven] = NOT_GIVEN,
+        audio: Union[Any, NotGiven] = NOT_GIVEN,
+        max_completion_tokens: Union[int, NotGiven] = NOT_GIVEN,
+        metadata: Union[Dict[str, str], NotGiven] = NOT_GIVEN,
+        modalities: Union[List[Any], NotGiven] = NOT_GIVEN,
+        prediction: Union[Any, NotGiven] = NOT_GIVEN,
+        reasoning_effort: Union[Any, NotGiven] = NOT_GIVEN,
+        store: Union[bool, NotGiven] = NOT_GIVEN,
         **kwargs: Any,
     ) -> ParsedChatCompletion:
         response = await self.openai_client.beta.chat.completions.parse(
@@ -89,6 +124,13 @@ async def parse(
             model=model,
             response_format=response_format,
             tools=tools,
+            audio=audio,
+            max_completion_tokens=max_completion_tokens,
+            metadata=metadata,
+            modalities=modalities,
+            prediction=prediction,
+            reasoning_effort=reasoning_effort,
+            store=store,
             extra_body=kwargs,
         )
         return response  # type: ignore [return-value]
@@ -100,6 +142,13 @@ async def stream(
         model: Union[str, Any] = "portkey-default",
         response_format: Union[Any, NotGiven] = NOT_GIVEN,
         tools: Union[Any, NotGiven] = NOT_GIVEN,
+        audio: Union[Any, NotGiven] = NOT_GIVEN,
+        max_completion_tokens: Union[int, NotGiven] = NOT_GIVEN,
+        metadata: Union[Dict[str, str], NotGiven] = NOT_GIVEN,
+        modalities: Union[List[Any], NotGiven] = NOT_GIVEN,
+        prediction: Union[Any, NotGiven] = NOT_GIVEN,
+        reasoning_effort: Union[Any, NotGiven] = NOT_GIVEN,
+        store: Union[bool, NotGiven] = NOT_GIVEN,
         **kwargs: Any,
     ) -> Any:
         async with self.openai_client.beta.chat.completions.stream(
@@ -107,6 +156,13 @@ async def stream(
             model=model,
             response_format=response_format,
             tools=tools,
+            audio=audio,
+            max_completion_tokens=max_completion_tokens,
+            metadata=metadata,
+            modalities=modalities,
+            prediction=prediction,
+            reasoning_effort=reasoning_effort,
+            store=store,
             extra_body=kwargs,
         ) as stream:
             async for event in stream:
diff --git a/portkey_ai/api_resources/apis/beta_realtime.py b/portkey_ai/api_resources/apis/beta_realtime.py
new file mode 100644
index 00000000..3ddcd6bc
--- /dev/null
+++ b/portkey_ai/api_resources/apis/beta_realtime.py
@@ -0,0 +1,137 @@
+import json
+from typing import Any, Iterable, List, Union
+from portkey_ai._vendor.openai.resources.beta.realtime.realtime import (
+    AsyncRealtimeConnectionManager,
+    RealtimeConnectionManager,
+)
+from portkey_ai._vendor.openai.types.websocket_connection_options import (
+    WebsocketConnectionOptions,
+)
+from portkey_ai.api_resources.apis.api_resource import APIResource, AsyncAPIResource
+from portkey_ai.api_resources.client import AsyncPortkey, Portkey
+from portkey_ai.api_resources.types.beta_realtime import SessionCreateResponse
+from ..._vendor.openai._types import NotGiven, NOT_GIVEN
+
+
+class BetaRealtime(APIResource):
+    def __init__(self, client: Portkey) -> None:
+        super().__init__(client)
+        self.openai_client = client.openai_client
+        self.sessions = BetaSessions(client)
+
+    def connect(
+        self,
+        *,
+        model: str,
+        websocket_connection_options: WebsocketConnectionOptions = {},
+        **kwargs,
+    ) -> RealtimeConnectionManager:
+        return self.openai_client.beta.realtime.connect(
+            model=model,
+            websocket_connection_options=websocket_connection_options,
+            extra_headers=self.openai_client.default_headers,
+            **kwargs,
+        )
+
+
+class AsyncBetaRealtime(AsyncAPIResource):
+    def __init__(self, client: AsyncPortkey) -> None:
+        super().__init__(client)
+        self.openai_client = client.openai_client
+        self.sessions = AsyncBetaSessions(client)
+
+    def connect(
+        self,
+        *,
+        model: str,
+        websocket_connection_options: WebsocketConnectionOptions = {},
+        **kwargs,
+    ) -> AsyncRealtimeConnectionManager:
+        return self.openai_client.beta.realtime.connect(
+            model=model,
+            websocket_connection_options=websocket_connection_options,
+            extra_headers=self.openai_client.default_headers,
+            **kwargs,
+        )
+
+
+class BetaSessions(APIResource):
+    def __init__(self, client: Portkey) -> None:
+        super().__init__(client)
+        self.openai_client = client.openai_client
+
+    def create(
+        self,
+        *,
+        model: Any = "portkey-default",
+        input_audio_format: Union[Any, NotGiven] = NOT_GIVEN,
+        input_audio_transcription: Union[Any, NotGiven] = NOT_GIVEN,
+        instructions: Union[str, NotGiven] = NOT_GIVEN,
+        max_response_output_tokens: Union[int, Any, NotGiven] = NOT_GIVEN,
+        modalities: Union[List[Any], NotGiven] = NOT_GIVEN,
+        output_audio_format: Union[Any, NotGiven] = NOT_GIVEN,
+        temperature: Union[float, NotGiven] = NOT_GIVEN,
+        tool_choice: Union[str, NotGiven] = NOT_GIVEN,
+        tools: Union[Iterable[Any], NotGiven] = NOT_GIVEN,
+        turn_detection: Union[Any, NotGiven] = NOT_GIVEN,
+        voice: Union[Any, NotGiven] = NOT_GIVEN,
+    ) -> SessionCreateResponse:
+        response = self.openai_client.with_raw_response.beta.realtime.sessions.create(
+            model=model,
+            input_audio_format=input_audio_format,
+            input_audio_transcription=input_audio_transcription,
+            instructions=instructions,
+            max_response_output_tokens=max_response_output_tokens,
+            modalities=modalities,
+            output_audio_format=output_audio_format,
+            temperature=temperature,
+            tool_choice=tool_choice,
+            tools=tools,
+            turn_detection=turn_detection,
+            voice=voice,
+        )
+        data = SessionCreateResponse(**json.loads(response.text))
+        data._headers = response.headers
+        return data
+
+
+class AsyncBetaSessions(AsyncAPIResource):
+    def __init__(self, client: AsyncPortkey) -> None:
+        super().__init__(client)
+        self.openai_client = client.openai_client
+
+    async def create(
+        self,
+        *,
+        model: Any = "portkey-default",
+        input_audio_format: Union[Any, NotGiven] = NOT_GIVEN,
+        input_audio_transcription: Union[Any, NotGiven] = NOT_GIVEN,
+        instructions: Union[str, NotGiven] = NOT_GIVEN,
+        max_response_output_tokens: Union[int, Any, NotGiven] = NOT_GIVEN,
+        modalities: Union[List[Any], NotGiven] = NOT_GIVEN,
+        output_audio_format: Union[Any, NotGiven] = NOT_GIVEN,
+        temperature: Union[float, NotGiven] = NOT_GIVEN,
+        tool_choice: Union[str, NotGiven] = NOT_GIVEN,
+        tools: Union[Iterable[Any], NotGiven] = NOT_GIVEN,
+        turn_detection: Union[Any, NotGiven] = NOT_GIVEN,
+        voice: Union[Any, NotGiven] = NOT_GIVEN,
+    ) -> SessionCreateResponse:
+        response = (
+            await self.openai_client.with_raw_response.beta.realtime.sessions.create(
+                model=model,
+                input_audio_format=input_audio_format,
+                input_audio_transcription=input_audio_transcription,
+                instructions=instructions,
+                max_response_output_tokens=max_response_output_tokens,
+                modalities=modalities,
+                output_audio_format=output_audio_format,
+                temperature=temperature,
+                tool_choice=tool_choice,
+                tools=tools,
+                turn_detection=turn_detection,
+                voice=voice,
+            )
+        )
+        data = SessionCreateResponse(**json.loads(response.text))
+        data._headers = response.headers
+        return data
diff --git a/portkey_ai/api_resources/apis/chat_complete.py b/portkey_ai/api_resources/apis/chat_complete.py
index 9579ea3f..cd899215 100644
--- a/portkey_ai/api_resources/apis/chat_complete.py
+++ b/portkey_ai/api_resources/apis/chat_complete.py
@@ -4,8 +4,10 @@
 from typing import (
     Any,
     AsyncIterator,
+    Dict,
     Iterable,
     Iterator,
+    List,
     Mapping,
     Optional,
     Union,
@@ -44,7 +46,21 @@ def __init__(self, client: Portkey) -> None:
         self.openai_client = client.openai_client
 
     def stream_create(  # type: ignore[return]
-        self, model, messages, stream, temperature, max_tokens, top_p, **kwargs
+        self,
+        model,
+        messages,
+        stream,
+        temperature,
+        max_tokens,
+        top_p,
+        audio,
+        max_completion_tokens,
+        metadata,
+        modalities,
+        prediction,
+        reasoning_effort,
+        store,
+        **kwargs,
     ) -> Union[ChatCompletions, Iterator[ChatCompletionChunk]]:
         with self.openai_client.with_streaming_response.chat.completions.create(
             model=model,
@@ -53,6 +69,13 @@ def stream_create(  # type: ignore[return]
             temperature=temperature,
             max_tokens=max_tokens,
             top_p=top_p,
+            audio=audio,
+            max_completion_tokens=max_completion_tokens,
+            metadata=metadata,
+            modalities=modalities,
+            prediction=prediction,
+            reasoning_effort=reasoning_effort,
+            store=store,
             extra_body=kwargs,
         ) as response:
             for line in response.iter_lines():
@@ -70,7 +93,21 @@ def stream_create(  # type: ignore[return]
                     return ""
 
     def normal_create(
-        self, model, messages, stream, temperature, max_tokens, top_p, **kwargs
+        self,
+        model,
+        messages,
+        stream,
+        temperature,
+        max_tokens,
+        top_p,
+        audio,
+        max_completion_tokens,
+        metadata,
+        modalities,
+        prediction,
+        reasoning_effort,
+        store,
+        **kwargs,
     ) -> ChatCompletions:
         response = self.openai_client.with_raw_response.chat.completions.create(
             model=model,
@@ -79,6 +116,13 @@ def normal_create(
             temperature=temperature,
             max_tokens=max_tokens,
             top_p=top_p,
+            audio=audio,
+            max_completion_tokens=max_completion_tokens,
+            metadata=metadata,
+            modalities=modalities,
+            prediction=prediction,
+            reasoning_effort=reasoning_effort,
+            store=store,
             extra_body=kwargs,
         )
         data = ChatCompletions(**json.loads(response.text))
@@ -94,6 +138,13 @@ def create(
         temperature: Union[float, NotGiven] = NOT_GIVEN,
         max_tokens: Union[int, NotGiven] = NOT_GIVEN,
         top_p: Union[float, NotGiven] = NOT_GIVEN,
+        audio: Optional[Any] = NOT_GIVEN,
+        max_completion_tokens: Union[int, NotGiven] = NOT_GIVEN,
+        metadata: Union[Dict[str, str], NotGiven] = NOT_GIVEN,
+        modalities: Union[List[Any], NotGiven] = NOT_GIVEN,
+        prediction: Union[Any, NotGiven] = NOT_GIVEN,
+        reasoning_effort: Union[Any, NotGiven] = NOT_GIVEN,
+        store: Union[Optional[bool], NotGiven] = NOT_GIVEN,
         **kwargs,
     ) -> Union[ChatCompletions, Iterator[ChatCompletionChunk]]:
         if stream is True:
@@ -104,6 +155,13 @@ def create(
                 temperature=temperature,
                 max_tokens=max_tokens,
                 top_p=top_p,
+                audio=audio,
+                max_completion_tokens=max_completion_tokens,
+                metadata=metadata,
+                modalities=modalities,
+                prediction=prediction,
+                reasoning_effort=reasoning_effort,
+                store=store,
                 **kwargs,
             )
         else:
@@ -114,6 +172,13 @@ def create(
                 temperature=temperature,
                 max_tokens=max_tokens,
                 top_p=top_p,
+                audio=audio,
+                max_completion_tokens=max_completion_tokens,
+                metadata=metadata,
+                modalities=modalities,
+                prediction=prediction,
+                reasoning_effort=reasoning_effort,
+                store=store,
                 **kwargs,
             )
 
@@ -124,7 +189,21 @@ def __init__(self, client: AsyncPortkey) -> None:
         self.openai_client = client.openai_client
 
     async def stream_create(
-        self, model, messages, stream, temperature, max_tokens, top_p, **kwargs
+        self,
+        model,
+        messages,
+        stream,
+        temperature,
+        max_tokens,
+        top_p,
+        audio,
+        max_completion_tokens,
+        metadata,
+        modalities,
+        prediction,
+        reasoning_effort,
+        store,
+        **kwargs,
     ) -> Union[ChatCompletions, AsyncIterator[ChatCompletionChunk]]:
         async with self.openai_client.with_streaming_response.chat.completions.create(
             model=model,
@@ -133,6 +212,13 @@ async def stream_create(
             temperature=temperature,
             max_tokens=max_tokens,
             top_p=top_p,
+            audio=audio,
+            max_completion_tokens=max_completion_tokens,
+            metadata=metadata,
+            modalities=modalities,
+            prediction=prediction,
+            reasoning_effort=reasoning_effort,
+            store=store,
             extra_body=kwargs,
         ) as response:
             async for line in response.iter_lines():
@@ -150,7 +236,21 @@ async def stream_create(
                     pass
 
     async def normal_create(
-        self, model, messages, stream, temperature, max_tokens, top_p, **kwargs
+        self,
+        model,
+        messages,
+        stream,
+        temperature,
+        max_tokens,
+        top_p,
+        audio,
+        max_completion_tokens,
+        metadata,
+        modalities,
+        prediction,
+        reasoning_effort,
+        store,
+        **kwargs,
     ) -> ChatCompletions:
         response = await self.openai_client.with_raw_response.chat.completions.create(
             model=model,
@@ -159,6 +259,13 @@ async def normal_create(
             temperature=temperature,
             max_tokens=max_tokens,
             top_p=top_p,
+            audio=audio,
+            max_completion_tokens=max_completion_tokens,
+            metadata=metadata,
+            modalities=modalities,
+            prediction=prediction,
+            reasoning_effort=reasoning_effort,
+            store=store,
             extra_body=kwargs,
         )
         data = ChatCompletions(**json.loads(response.text))
@@ -174,6 +281,13 @@ async def create(
         temperature: Union[float, NotGiven] = NOT_GIVEN,
         max_tokens: Union[int, NotGiven] = NOT_GIVEN,
         top_p: Union[float, NotGiven] = NOT_GIVEN,
+        audio: Optional[Any] = NOT_GIVEN,
+        max_completion_tokens: Union[int, NotGiven] = NOT_GIVEN,
+        metadata: Union[Dict[str, str], NotGiven] = NOT_GIVEN,
+        modalities: Union[List[Any], NotGiven] = NOT_GIVEN,
+        prediction: Union[Any, NotGiven] = NOT_GIVEN,
+        reasoning_effort: Union[Any, NotGiven] = NOT_GIVEN,
+        store: Union[Optional[bool], NotGiven] = NOT_GIVEN,
         **kwargs,
     ) -> Union[ChatCompletions, AsyncIterator[ChatCompletionChunk]]:
         if stream is True:
@@ -184,6 +298,13 @@ async def create(
                 temperature=temperature,
                 max_tokens=max_tokens,
                 top_p=top_p,
+                audio=audio,
+                max_completion_tokens=max_completion_tokens,
+                metadata=metadata,
+                modalities=modalities,
+                prediction=prediction,
+                reasoning_effort=reasoning_effort,
+                store=store,
                 **kwargs,
             )
         else:
@@ -194,6 +315,13 @@ async def create(
                 temperature=temperature,
                 max_tokens=max_tokens,
                 top_p=top_p,
+                audio=audio,
+                max_completion_tokens=max_completion_tokens,
+                metadata=metadata,
+                modalities=modalities,
+                prediction=prediction,
+                reasoning_effort=reasoning_effort,
+                store=store,
                 **kwargs,
             )
 
diff --git a/portkey_ai/api_resources/apis/complete.py b/portkey_ai/api_resources/apis/complete.py
index 00d9e903..f0a1cbad 100644
--- a/portkey_ai/api_resources/apis/complete.py
+++ b/portkey_ai/api_resources/apis/complete.py
@@ -1,5 +1,5 @@
 import json
-from typing import AsyncIterator, Iterator, Optional, Union
+from typing import Any, AsyncIterator, Dict, Iterable, Iterator, List, Optional, Union
 from portkey_ai.api_resources.client import AsyncPortkey, Portkey
 from ..._vendor.openai._types import NotGiven, NOT_GIVEN
 
@@ -18,7 +18,26 @@ def __init__(self, client: Portkey) -> None:
         self.client = client
 
     def stream_create(  # type: ignore[return]
-        self, model, prompt, stream, temperature, max_tokens, top_p, **kwargs
+        self,
+        model,
+        prompt,
+        stream,
+        temperature,
+        max_tokens,
+        top_p,
+        best_of,
+        echo,
+        frequency_penalty,
+        logit_bias,
+        logprobs,
+        n,
+        presence_penalty,
+        seed,
+        stop,
+        suffix,
+        user,
+        stream_options,
+        **kwargs,
     ) -> Union[TextCompletion, Iterator[TextCompletionChunk]]:
         with self.openai_client.with_streaming_response.completions.create(
             model=model,
@@ -27,6 +46,18 @@ def stream_create(  # type: ignore[return]
             temperature=temperature,
             max_tokens=max_tokens,
             top_p=top_p,
+            best_of=best_of,
+            echo=echo,
+            frequency_penalty=frequency_penalty,
+            logit_bias=logit_bias,
+            logprobs=logprobs,
+            n=n,
+            presence_penalty=presence_penalty,
+            seed=seed,
+            stop=stop,
+            suffix=suffix,
+            user=user,
+            stream_options=stream_options,
             extra_body=kwargs,
         ) as response:
             for line in response.iter_lines():
@@ -44,7 +75,26 @@ def stream_create(  # type: ignore[return]
                     return ""
 
     def normal_create(
-        self, model, prompt, stream, temperature, max_tokens, top_p, **kwargs
+        self,
+        model,
+        prompt,
+        stream,
+        temperature,
+        max_tokens,
+        top_p,
+        best_of,
+        echo,
+        frequency_penalty,
+        logit_bias,
+        logprobs,
+        n,
+        presence_penalty,
+        seed,
+        stop,
+        suffix,
+        user,
+        stream_options,
+        **kwargs,
     ) -> TextCompletion:
         response = self.openai_client.with_raw_response.completions.create(
             model=model,
@@ -53,6 +103,18 @@ def normal_create(
             temperature=temperature,
             max_tokens=max_tokens,
             top_p=top_p,
+            best_of=best_of,
+            echo=echo,
+            frequency_penalty=frequency_penalty,
+            logit_bias=logit_bias,
+            logprobs=logprobs,
+            n=n,
+            presence_penalty=presence_penalty,
+            seed=seed,
+            stop=stop,
+            suffix=suffix,
+            user=user,
+            stream_options=stream_options,
             extra_body=kwargs,
         )
         data = TextCompletion(**json.loads(response.text))
@@ -63,11 +125,25 @@ def create(
         self,
         *,
         model: Optional[str] = "portkey-default",
-        prompt: Optional[str] = None,
+        prompt: Union[
+            str, List[str], Iterable[int], Iterable[Iterable[int]], None
+        ] = None,
         stream: Union[bool, NotGiven] = NOT_GIVEN,
         temperature: Union[float, NotGiven] = NOT_GIVEN,
         max_tokens: Union[int, NotGiven] = NOT_GIVEN,
         top_p: Union[bool, NotGiven] = NOT_GIVEN,
+        best_of: Union[int, NotGiven] = NOT_GIVEN,
+        echo: Union[bool, NotGiven] = NOT_GIVEN,
+        frequency_penalty: Union[float, NotGiven] = NOT_GIVEN,
+        logit_bias: Union[Dict[str, int], NotGiven] = NOT_GIVEN,
+        logprobs: Union[int, NotGiven] = NOT_GIVEN,
+        n: Union[int, NotGiven] = NOT_GIVEN,
+        presence_penalty: Union[float, NotGiven] = NOT_GIVEN,
+        seed: Union[int, NotGiven] = NOT_GIVEN,
+        stop: Union[Optional[str], List[str], None, NotGiven] = NOT_GIVEN,
+        suffix: Union[str, NotGiven] = NOT_GIVEN,
+        user: Union[str, NotGiven] = NOT_GIVEN,
+        stream_options: Optional[Any] = NOT_GIVEN,
         **kwargs,
     ) -> Union[TextCompletion, Iterator[TextCompletionChunk]]:
         if stream is True:
@@ -78,6 +154,18 @@ def create(
                 temperature=temperature,
                 max_tokens=max_tokens,
                 top_p=top_p,
+                best_of=best_of,
+                echo=echo,
+                frequency_penalty=frequency_penalty,
+                logit_bias=logit_bias,
+                logprobs=logprobs,
+                n=n,
+                presence_penalty=presence_penalty,
+                seed=seed,
+                stop=stop,
+                suffix=suffix,
+                user=user,
+                stream_options=stream_options,
                 **kwargs,
             )
         else:
@@ -88,6 +176,18 @@ def create(
                 temperature=temperature,
                 max_tokens=max_tokens,
                 top_p=top_p,
+                best_of=best_of,
+                echo=echo,
+                frequency_penalty=frequency_penalty,
+                logit_bias=logit_bias,
+                logprobs=logprobs,
+                n=n,
+                presence_penalty=presence_penalty,
+                seed=seed,
+                stop=stop,
+                suffix=suffix,
+                user=user,
+                stream_options=stream_options,
                 **kwargs,
             )
 
@@ -98,7 +198,26 @@ def __init__(self, client: AsyncPortkey) -> None:
         self.openai_client = client.openai_client
 
     async def stream_create(
-        self, model, prompt, stream, temperature, max_tokens, top_p, **kwargs
+        self,
+        model,
+        prompt,
+        stream,
+        temperature,
+        max_tokens,
+        top_p,
+        best_of,
+        echo,
+        frequency_penalty,
+        logit_bias,
+        logprobs,
+        n,
+        presence_penalty,
+        seed,
+        stop,
+        suffix,
+        user,
+        stream_options,
+        **kwargs,
     ) -> Union[TextCompletion, AsyncIterator[TextCompletionChunk]]:
         async with self.openai_client.with_streaming_response.completions.create(
             model=model,
@@ -107,6 +226,18 @@ async def stream_create(
             temperature=temperature,
             max_tokens=max_tokens,
             top_p=top_p,
+            best_of=best_of,
+            echo=echo,
+            frequency_penalty=frequency_penalty,
+            logit_bias=logit_bias,
+            logprobs=logprobs,
+            n=n,
+            presence_penalty=presence_penalty,
+            seed=seed,
+            stop=stop,
+            suffix=suffix,
+            user=user,
+            stream_options=stream_options,
             extra_body=kwargs,
         ) as response:
             async for line in response.iter_lines():
@@ -124,7 +255,26 @@ async def stream_create(
                     pass
 
     async def normal_create(
-        self, model, prompt, stream, temperature, max_tokens, top_p, **kwargs
+        self,
+        model,
+        prompt,
+        stream,
+        temperature,
+        max_tokens,
+        top_p,
+        best_of,
+        echo,
+        frequency_penalty,
+        logit_bias,
+        logprobs,
+        n,
+        presence_penalty,
+        seed,
+        stop,
+        suffix,
+        user,
+        stream_options,
+        **kwargs,
     ) -> TextCompletion:
         response = await self.openai_client.with_raw_response.completions.create(
             model=model,
@@ -133,6 +283,18 @@ async def normal_create(
             temperature=temperature,
             max_tokens=max_tokens,
             top_p=top_p,
+            best_of=best_of,
+            echo=echo,
+            frequency_penalty=frequency_penalty,
+            logit_bias=logit_bias,
+            logprobs=logprobs,
+            n=n,
+            presence_penalty=presence_penalty,
+            seed=seed,
+            stop=stop,
+            suffix=suffix,
+            user=user,
+            stream_options=stream_options,
             extra_body=kwargs,
         )
         data = TextCompletion(**json.loads(response.text))
@@ -143,11 +305,25 @@ async def create(
         self,
         *,
         model: Optional[str] = "portkey-default",
-        prompt: Optional[str] = None,
+        prompt: Union[
+            str, List[str], Iterable[int], Iterable[Iterable[int]], None
+        ] = None,
         stream: Union[bool, NotGiven] = NOT_GIVEN,
         temperature: Union[float, NotGiven] = NOT_GIVEN,
         max_tokens: Union[int, NotGiven] = NOT_GIVEN,
         top_p: Union[bool, NotGiven] = NOT_GIVEN,
+        best_of: Union[int, NotGiven] = NOT_GIVEN,
+        echo: Union[bool, NotGiven] = NOT_GIVEN,
+        frequency_penalty: Union[float, NotGiven] = NOT_GIVEN,
+        logit_bias: Union[Dict[str, int], NotGiven] = NOT_GIVEN,
+        logprobs: Union[int, NotGiven] = NOT_GIVEN,
+        n: Union[int, NotGiven] = NOT_GIVEN,
+        presence_penalty: Union[float, NotGiven] = NOT_GIVEN,
+        seed: Union[int, NotGiven] = NOT_GIVEN,
+        stop: Union[Optional[str], List[str], None, NotGiven] = NOT_GIVEN,
+        suffix: Union[str, NotGiven] = NOT_GIVEN,
+        user: Union[str, NotGiven] = NOT_GIVEN,
+        stream_options: Optional[Any] = NOT_GIVEN,
         **kwargs,
     ) -> Union[TextCompletion, AsyncIterator[TextCompletionChunk]]:
         if stream is True:
@@ -158,6 +334,18 @@ async def create(
                 temperature=temperature,
                 max_tokens=max_tokens,
                 top_p=top_p,
+                best_of=best_of,
+                echo=echo,
+                frequency_penalty=frequency_penalty,
+                logit_bias=logit_bias,
+                logprobs=logprobs,
+                n=n,
+                presence_penalty=presence_penalty,
+                seed=seed,
+                stop=stop,
+                suffix=suffix,
+                user=user,
+                stream_options=stream_options,
                 **kwargs,
             )
         else:
@@ -168,5 +356,17 @@ async def create(
                 temperature=temperature,
                 max_tokens=max_tokens,
                 top_p=top_p,
+                best_of=best_of,
+                echo=echo,
+                frequency_penalty=frequency_penalty,
+                logit_bias=logit_bias,
+                logprobs=logprobs,
+                n=n,
+                presence_penalty=presence_penalty,
+                seed=seed,
+                stop=stop,
+                suffix=suffix,
+                user=user,
+                stream_options=stream_options,
                 **kwargs,
             )
diff --git a/portkey_ai/api_resources/apis/fine_tuning.py b/portkey_ai/api_resources/apis/fine_tuning.py
index 237ca92a..0b12f4e1 100644
--- a/portkey_ai/api_resources/apis/fine_tuning.py
+++ b/portkey_ai/api_resources/apis/fine_tuning.py
@@ -35,6 +35,7 @@ def create(
         integrations: Union[
             Optional[Iterable[job_create_params.Integration]], NotGiven
         ] = NOT_GIVEN,
+        method: Union[job_create_params.Method, NotGiven] = NOT_GIVEN,
         seed: Union[Optional[int], NotGiven] = NOT_GIVEN,
         suffix: Union[Optional[str], NotGiven] = NOT_GIVEN,
         validation_file: Union[Optional[str], NotGiven] = NOT_GIVEN,
@@ -45,6 +46,7 @@ def create(
             training_file=training_file,
             hyperparameters=hyperparameters,
             integrations=integrations,
+            method=method,
             seed=seed,
             suffix=suffix,
             validation_file=validation_file,
@@ -163,6 +165,7 @@ async def create(
         integrations: Union[
             Optional[Iterable[job_create_params.Integration]], NotGiven
         ] = NOT_GIVEN,
+        method: Union[job_create_params.Method, NotGiven] = NOT_GIVEN,
         seed: Union[Optional[int], NotGiven] = NOT_GIVEN,
         suffix: Union[Optional[str], NotGiven] = NOT_GIVEN,
         validation_file: Union[Optional[str], NotGiven] = NOT_GIVEN,
@@ -173,6 +176,7 @@ async def create(
             training_file=training_file,
             hyperparameters=hyperparameters,
             integrations=integrations,
+            method=method,
             seed=seed,
             suffix=suffix,
             validation_file=validation_file,
diff --git a/portkey_ai/api_resources/apis/images.py b/portkey_ai/api_resources/apis/images.py
index 7e8092f0..c9a22660 100644
--- a/portkey_ai/api_resources/apis/images.py
+++ b/portkey_ai/api_resources/apis/images.py
@@ -13,27 +13,23 @@ def __init__(self, client: Portkey) -> None:
         self.openai_client = client.openai_client
 
     @typing.no_type_check
-    def generate(
+    def create_variation(
         self,
         *,
-        prompt: str,
-        model: Union[str, NotGiven] = NOT_GIVEN,
+        image,
         n: Union[int, NotGiven] = NOT_GIVEN,
-        quality: Union[str, NotGiven] = NOT_GIVEN,
+        model: Union[str, NotGiven] = NOT_GIVEN,
         response_format: Union[str, NotGiven] = NOT_GIVEN,
         size: Union[str, NotGiven] = NOT_GIVEN,
         user: Union[str, NotGiven] = NOT_GIVEN,
-        style: Union[str, NotGiven] = NOT_GIVEN,
         **kwargs
     ) -> ImagesResponse:
-        response = self.openai_client.with_raw_response.images.generate(
-            prompt=prompt,
-            model=model,
+        response = self.openai_client.with_raw_response.images.create_variation(
+            image=image,
             n=n,
-            quality=quality,
+            model=model,
             response_format=response_format,
             size=size,
-            style=style,
             user=user,
             extra_body=kwargs,
         )
@@ -73,21 +69,27 @@ def edit(
         return data
 
     @typing.no_type_check
-    def create_variation(
+    def generate(
         self,
         *,
-        image,
+        prompt: str,
+        model: Union[str, NotGiven] = NOT_GIVEN,
         n: Union[int, NotGiven] = NOT_GIVEN,
+        quality: Union[str, NotGiven] = NOT_GIVEN,
         response_format: Union[str, NotGiven] = NOT_GIVEN,
         size: Union[str, NotGiven] = NOT_GIVEN,
         user: Union[str, NotGiven] = NOT_GIVEN,
+        style: Union[str, NotGiven] = NOT_GIVEN,
         **kwargs
     ) -> ImagesResponse:
-        response = self.openai_client.with_raw_response.images.create_variation(
-            image=image,
+        response = self.openai_client.with_raw_response.images.generate(
+            prompt=prompt,
+            model=model,
             n=n,
+            quality=quality,
             response_format=response_format,
             size=size,
+            style=style,
             user=user,
             extra_body=kwargs,
         )
@@ -103,27 +105,21 @@ def __init__(self, client: AsyncPortkey) -> None:
         self.openai_client = client.openai_client
 
     @typing.no_type_check
-    async def generate(
+    async def create_variation(
         self,
         *,
-        prompt: str,
-        model: Union[str, NotGiven] = NOT_GIVEN,
+        image,
         n: Union[int, NotGiven] = NOT_GIVEN,
-        quality: Union[str, NotGiven] = NOT_GIVEN,
         response_format: Union[str, NotGiven] = NOT_GIVEN,
         size: Union[str, NotGiven] = NOT_GIVEN,
         user: Union[str, NotGiven] = NOT_GIVEN,
-        style: Union[str, NotGiven] = NOT_GIVEN,
         **kwargs
     ) -> ImagesResponse:
-        response = await self.openai_client.with_raw_response.images.generate(
-            prompt=prompt,
-            model=model,
+        response = await self.openai_client.with_raw_response.images.create_variation(
+            image=image,
             n=n,
-            quality=quality,
             response_format=response_format,
             size=size,
-            style=style,
             user=user,
             extra_body=kwargs,
         )
@@ -163,21 +159,27 @@ async def edit(
         return data
 
     @typing.no_type_check
-    async def create_variation(
+    async def generate(
         self,
         *,
-        image,
+        prompt: str,
+        model: Union[str, NotGiven] = NOT_GIVEN,
         n: Union[int, NotGiven] = NOT_GIVEN,
+        quality: Union[str, NotGiven] = NOT_GIVEN,
         response_format: Union[str, NotGiven] = NOT_GIVEN,
         size: Union[str, NotGiven] = NOT_GIVEN,
         user: Union[str, NotGiven] = NOT_GIVEN,
+        style: Union[str, NotGiven] = NOT_GIVEN,
         **kwargs
     ) -> ImagesResponse:
-        response = await self.openai_client.with_raw_response.images.create_variation(
-            image=image,
+        response = await self.openai_client.with_raw_response.images.generate(
+            prompt=prompt,
+            model=model,
             n=n,
+            quality=quality,
             response_format=response_format,
             size=size,
+            style=style,
             user=user,
             extra_body=kwargs,
         )
diff --git a/portkey_ai/api_resources/apis/main_files.py b/portkey_ai/api_resources/apis/main_files.py
index f5063efd..33449e27 100644
--- a/portkey_ai/api_resources/apis/main_files.py
+++ b/portkey_ai/api_resources/apis/main_files.py
@@ -1,5 +1,6 @@
 import json
 from typing import Any, Union
+import typing_extensions
 from portkey_ai._vendor.openai._types import NOT_GIVEN, NotGiven
 from portkey_ai.api_resources.apis.api_resource import APIResource, AsyncAPIResource
 from portkey_ai.api_resources.client import AsyncPortkey, Portkey
@@ -25,15 +26,6 @@ def create(self, file, purpose, **kwargs) -> FileObject:
 
         return data
 
-    def list(self, purpose: Union[str, NotGiven] = NOT_GIVEN, **kwargs) -> FileList:
-        response = self.openai_client.with_raw_response.files.list(
-            purpose=purpose, **kwargs
-        )
-        data = FileList(**json.loads(response.text))
-        data._headers = response.headers
-
-        return data
-
     def retrieve(self, file_id, **kwargs) -> FileObject:
         if kwargs:
             response = self.openai_client.with_raw_response.files.retrieve(
@@ -48,6 +40,27 @@ def retrieve(self, file_id, **kwargs) -> FileObject:
 
         return data
 
+    def list(
+        self,
+        *,
+        purpose: Union[str, NotGiven] = NOT_GIVEN,
+        after: Union[str, NotGiven] = NOT_GIVEN,
+        limit: Union[int, NotGiven] = NOT_GIVEN,
+        order: Union[Any, NotGiven] = NOT_GIVEN,
+        **kwargs,
+    ) -> FileList:
+        response = self.openai_client.with_raw_response.files.list(
+            purpose=purpose,
+            after=after,
+            limit=limit,
+            order=order,
+            **kwargs,
+        )
+        data = FileList(**json.loads(response.text))
+        data._headers = response.headers
+
+        return data
+
     def delete(self, file_id, **kwargs) -> FileDeleted:
         response = self.openai_client.with_raw_response.files.delete(
             file_id=file_id, extra_body=kwargs
@@ -66,6 +79,7 @@ def content(self, file_id, **kwargs) -> Any:
             response = self.openai_client.files.content(file_id=file_id)
         return response
 
+    @typing_extensions.deprecated("The `.content()` method should be used instead")
     def retrieve_content(self, file_id, **kwargs) -> Any:
         if kwargs:
             response = self.openai_client.files.content(
@@ -105,17 +119,6 @@ async def create(self, file, purpose, **kwargs) -> FileObject:
 
         return data
 
-    async def list(
-        self, purpose: Union[str, NotGiven] = NOT_GIVEN, **kwargs
-    ) -> FileList:
-        response = await self.openai_client.with_raw_response.files.list(
-            purpose=purpose, **kwargs
-        )
-        data = FileList(**json.loads(response.text))
-        data._headers = response.headers
-
-        return data
-
     async def retrieve(self, file_id, **kwargs) -> FileObject:
         if kwargs:
             response = await self.openai_client.with_raw_response.files.retrieve(
@@ -130,6 +133,27 @@ async def retrieve(self, file_id, **kwargs) -> FileObject:
 
         return data
 
+    async def list(
+        self,
+        *,
+        purpose: Union[str, NotGiven] = NOT_GIVEN,
+        after: Union[str, NotGiven] = NOT_GIVEN,
+        limit: Union[int, NotGiven] = NOT_GIVEN,
+        order: Union[Any, NotGiven] = NOT_GIVEN,
+        **kwargs,
+    ) -> FileList:
+        response = await self.openai_client.with_raw_response.files.list(
+            purpose=purpose,
+            after=after,
+            limit=limit,
+            order=order,
+            **kwargs,
+        )
+        data = FileList(**json.loads(response.text))
+        data._headers = response.headers
+
+        return data
+
     async def delete(self, file_id, **kwargs) -> FileDeleted:
         response = await self.openai_client.with_raw_response.files.delete(
             file_id=file_id, extra_body=kwargs
@@ -148,6 +172,7 @@ async def content(self, file_id, **kwargs) -> Any:
             response = await self.openai_client.files.content(file_id=file_id)
         return response
 
+    @typing_extensions.deprecated("The `.content()` method should be used instead")
     async def retrieve_content(self, file_id, **kwargs) -> Any:
         if kwargs:
             response = await self.openai_client.files.content(
diff --git a/portkey_ai/api_resources/apis/models.py b/portkey_ai/api_resources/apis/models.py
index 8b4e9322..333a13af 100644
--- a/portkey_ai/api_resources/apis/models.py
+++ b/portkey_ai/api_resources/apis/models.py
@@ -11,12 +11,6 @@ def __init__(self, client: Portkey) -> None:
         super().__init__(client)
         self.openai_client = client.openai_client
 
-    def list(self, **kwargs) -> ModelList:
-        response = self.openai_client.with_raw_response.models.list(**kwargs)
-        data = ModelList(**json.loads(response.text))
-        data._headers = response.headers
-        return data
-
     def retrieve(
         self, model: str, *, timeout: Union[float, NotGiven] = NOT_GIVEN, **kwargs
     ) -> Model:
@@ -32,6 +26,12 @@ def retrieve(
         data._headers = response.headers
         return data
 
+    def list(self, **kwargs) -> ModelList:
+        response = self.openai_client.with_raw_response.models.list(**kwargs)
+        data = ModelList(**json.loads(response.text))
+        data._headers = response.headers
+        return data
+
     def delete(
         self, model: str, *, timeout: Union[float, NotGiven] = NOT_GIVEN, **kwargs
     ) -> ModelDeleted:
@@ -48,12 +48,6 @@ def __init__(self, client: AsyncPortkey) -> None:
         super().__init__(client)
         self.openai_client = client.openai_client
 
-    async def list(self, **kwargs) -> ModelList:
-        response = await self.openai_client.with_raw_response.models.list(**kwargs)
-        data = ModelList(**json.loads(response.text))
-        data._headers = response.headers
-        return data
-
     async def retrieve(
         self, model: str, *, timeout: Union[float, NotGiven] = NOT_GIVEN, **kwargs
     ) -> Model:
@@ -69,6 +63,12 @@ async def retrieve(
         data._headers = response.headers
         return data
 
+    async def list(self, **kwargs) -> ModelList:
+        response = await self.openai_client.with_raw_response.models.list(**kwargs)
+        data = ModelList(**json.loads(response.text))
+        data._headers = response.headers
+        return data
+
     async def delete(
         self, model: str, *, timeout: Union[float, NotGiven] = NOT_GIVEN, **kwargs
     ) -> ModelDeleted:
diff --git a/portkey_ai/api_resources/apis/moderations.py b/portkey_ai/api_resources/apis/moderations.py
index 3abe5f9b..c099a4b9 100644
--- a/portkey_ai/api_resources/apis/moderations.py
+++ b/portkey_ai/api_resources/apis/moderations.py
@@ -1,5 +1,5 @@
 import json
-from typing import List, Union
+from typing import Any, Iterable, List, Union
 from portkey_ai.api_resources.apis.api_resource import APIResource, AsyncAPIResource
 from portkey_ai.api_resources.client import AsyncPortkey, Portkey
 from ..._vendor.openai._types import NotGiven, NOT_GIVEN
@@ -14,7 +14,7 @@ def __init__(self, client: Portkey) -> None:
     def create(
         self,
         *,
-        input: Union[str, List[str]],
+        input: Union[str, List[str], Iterable[Any]],
         model: Union[str, NotGiven] = NOT_GIVEN,
         **kwargs
     ) -> ModerationCreateResponse:
@@ -35,7 +35,7 @@ def __init__(self, client: AsyncPortkey) -> None:
     async def create(
         self,
         *,
-        input: Union[str, List[str]],
+        input: Union[str, List[str], Iterable[Any]],
         model: Union[str, NotGiven] = NOT_GIVEN,
         **kwargs
     ) -> ModerationCreateResponse:
diff --git a/portkey_ai/api_resources/apis/threads.py b/portkey_ai/api_resources/apis/threads.py
index 7c4c3607..1d4a101d 100644
--- a/portkey_ai/api_resources/apis/threads.py
+++ b/portkey_ai/api_resources/apis/threads.py
@@ -1,5 +1,14 @@
 import json
-from typing import Any, AsyncIterator, Iterable, Iterator, Literal, Optional, Union
+from typing import (
+    Any,
+    AsyncIterator,
+    Iterable,
+    Iterator,
+    List,
+    Literal,
+    Optional,
+    Union,
+)
 import typing
 
 from portkey_ai._vendor.openai.types.beta.assistant_stream_event import (
@@ -283,6 +292,38 @@ def create(
 
         return data
 
+    def retrieve(self, thread_id, message_id, **kwargs) -> ThreadMessage:
+        if kwargs:
+            response = (
+                self.openai_client.with_raw_response.beta.threads.messages.retrieve(
+                    thread_id=thread_id, message_id=message_id, extra_body=kwargs
+                )
+            )
+        else:
+            response = (
+                self.openai_client.with_raw_response.beta.threads.messages.retrieve(
+                    thread_id=thread_id, message_id=message_id
+                )
+            )
+        data = ThreadMessage(**json.loads(response.text))
+        data._headers = response.headers
+        return data
+
+    def update(
+        self,
+        thread_id,
+        message_id,
+        *,
+        metadata: Union[Optional[object], NotGiven] = NOT_GIVEN,
+        **kwargs,
+    ) -> ThreadMessage:
+        response = self.openai_client.with_raw_response.beta.threads.messages.update(
+            thread_id=thread_id, message_id=message_id, metadata=metadata, **kwargs
+        )
+        data = ThreadMessage(**json.loads(response.text))
+        data._headers = response.headers
+        return data
+
     def list(
         self,
         thread_id,
@@ -307,31 +348,6 @@ def list(
         data._headers = response.headers
         return data
 
-    def retrieve(self, thread_id, message_id, **kwargs) -> ThreadMessage:
-        if kwargs:
-            response = (
-                self.openai_client.with_raw_response.beta.threads.messages.retrieve(
-                    thread_id=thread_id, message_id=message_id, extra_body=kwargs
-                )
-            )
-        else:
-            response = (
-                self.openai_client.with_raw_response.beta.threads.messages.retrieve(
-                    thread_id=thread_id, message_id=message_id
-                )
-            )
-        data = ThreadMessage(**json.loads(response.text))
-        data._headers = response.headers
-        return data
-
-    def update(self, thread_id, message_id, **kwargs) -> ThreadMessage:
-        response = self.openai_client.with_raw_response.beta.threads.messages.update(
-            thread_id=thread_id, message_id=message_id, **kwargs
-        )
-        data = ThreadMessage(**json.loads(response.text))
-        data._headers = response.headers
-        return data
-
     def delete(
         self, message_id: str, *, thread_id: str, **kwargs
     ) -> ThreadMessageDeleted:
@@ -418,15 +434,6 @@ def retrieve(self, thread_id, run_id, **kwargs) -> Run:
 
         return data
 
-    def list(self, thread_id, **kwargs) -> RunList:
-        response = self.openai_client.with_raw_response.beta.threads.runs.list(
-            thread_id=thread_id, **kwargs
-        )
-        data = RunList(**json.loads(response.text))
-        data._headers = response.headers
-
-        return data
-
     def update(
         self,
         thread_id,
@@ -442,13 +449,25 @@ def update(
 
         return data
 
-    def submit_tool_outputs(self, thread_id, tool_outputs, run_id, **kwargs) -> Run:
-        response = (
-            self.openai_client.with_raw_response.beta.threads.runs.submit_tool_outputs(
-                thread_id=thread_id, run_id=run_id, tool_outputs=tool_outputs, **kwargs
-            )
+    def list(
+        self,
+        thread_id,
+        *,
+        after: Union[str, NotGiven] = NOT_GIVEN,
+        before: Union[str, NotGiven] = NOT_GIVEN,
+        limit: Union[int, NotGiven] = NOT_GIVEN,
+        order: Union[NotGiven, Literal["asc", "desc"]] = NOT_GIVEN,
+        **kwargs,
+    ) -> RunList:
+        response = self.openai_client.with_raw_response.beta.threads.runs.list(
+            thread_id=thread_id,
+            after=after,
+            before=before,
+            limit=limit,
+            order=order,
+            **kwargs,
         )
-        data = Run(**json.loads(response.text))
+        data = RunList(**json.loads(response.text))
         data._headers = response.headers
 
         return data
@@ -466,6 +485,7 @@ def create_and_poll(
         self,
         *,
         assistant_id: str,
+        include: Union[List[Any], NotGiven] = NOT_GIVEN,
         additional_instructions: Union[Optional[str], NotGiven] = NOT_GIVEN,
         additional_messages: Union[
             Optional[Iterable[run_create_params.AdditionalMessage]], NotGiven
@@ -475,6 +495,7 @@ def create_and_poll(
         max_prompt_tokens: Union[Optional[int], NotGiven] = NOT_GIVEN,
         metadata: Union[Optional[object], NotGiven] = NOT_GIVEN,
         model: Union[str, None, NotGiven] = NOT_GIVEN,
+        parallel_tool_calls: Union[bool, NotGiven] = NOT_GIVEN,
         response_format: Union[
             Optional[AssistantResponseFormatOptionParam], NotGiven
         ] = NOT_GIVEN,
@@ -493,6 +514,7 @@ def create_and_poll(
     ) -> Run:
         response = self.openai_client.beta.threads.runs.create_and_poll(
             assistant_id=assistant_id,
+            include=include,
             additional_instructions=additional_instructions,
             additional_messages=additional_messages,
             instructions=instructions,
@@ -500,6 +522,7 @@ def create_and_poll(
             max_prompt_tokens=max_prompt_tokens,
             metadata=metadata,
             model=model,
+            parallel_tool_calls=parallel_tool_calls,
             response_format=response_format,
             temperature=temperature,
             tool_choice=tool_choice,
@@ -528,6 +551,7 @@ def create_and_stream(
         max_prompt_tokens: Union[Optional[int], NotGiven] = NOT_GIVEN,
         metadata: Union[Optional[object], NotGiven] = NOT_GIVEN,
         model: Union[str, None, NotGiven] = NOT_GIVEN,
+        parallel_tool_calls: Union[bool, NotGiven] = NOT_GIVEN,
         response_format: Union[
             Optional[AssistantResponseFormatOptionParam], NotGiven
         ] = NOT_GIVEN,
@@ -556,6 +580,7 @@ def create_and_stream(
             max_prompt_tokens=max_prompt_tokens,
             metadata=metadata,
             model=model,
+            parallel_tool_calls=parallel_tool_calls,
             response_format=response_format,
             temperature=temperature,
             tool_choice=tool_choice,
@@ -588,6 +613,7 @@ def stream(
         self,
         *,
         assistant_id: str,
+        include: Union[List[Any], NotGiven] = NOT_GIVEN,
         additional_instructions: Union[Optional[str], NotGiven] = NOT_GIVEN,
         additional_messages: Union[
             Optional[Iterable[run_create_params.AdditionalMessage]], NotGiven
@@ -597,6 +623,7 @@ def stream(
         max_prompt_tokens: Union[Optional[int], NotGiven] = NOT_GIVEN,
         metadata: Union[Optional[object], NotGiven] = NOT_GIVEN,
         model: Union[str, None, NotGiven] = NOT_GIVEN,
+        parallel_tool_calls: Union[bool, NotGiven] = NOT_GIVEN,
         response_format: Union[
             Optional[AssistantResponseFormatOptionParam], NotGiven
         ] = NOT_GIVEN,
@@ -618,6 +645,7 @@ def stream(
     ]:
         response = self.openai_client.beta.threads.runs.stream(
             assistant_id=assistant_id,
+            include=include,
             additional_instructions=additional_instructions,
             additional_messages=additional_messages,
             instructions=instructions,
@@ -625,6 +653,7 @@ def stream(
             max_prompt_tokens=max_prompt_tokens,
             metadata=metadata,
             model=model,
+            parallel_tool_calls=parallel_tool_calls,
             response_format=response_format,
             temperature=temperature,
             tool_choice=tool_choice,
@@ -638,6 +667,29 @@ def stream(
         data = response
         return data
 
+    def submit_tool_outputs(
+        self,
+        run_id,
+        *,
+        thread_id,
+        tool_outputs,
+        stream: Union[bool, NotGiven] = NOT_GIVEN,
+        **kwargs,
+    ) -> Run:
+        response = (
+            self.openai_client.with_raw_response.beta.threads.runs.submit_tool_outputs(
+                thread_id=thread_id,
+                run_id=run_id,
+                tool_outputs=tool_outputs,
+                stream=stream,
+                **kwargs,
+            )
+        )
+        data = Run(**json.loads(response.text))
+        data._headers = response.headers
+
+        return data
+
     def submit_tool_outputs_and_poll(
         self,
         *,
@@ -687,29 +739,31 @@ def __init__(self, client: Portkey) -> None:
         super().__init__(client)
         self.openai_client = client.openai_client
 
-    def list(self, thread_id, run_id, **kwargs) -> RunStepList:
-        response = self.openai_client.with_raw_response.beta.threads.runs.steps.list(
-            thread_id=thread_id, run_id=run_id, **kwargs
-        )
-        data = RunStepList(**json.loads(response.text))
-        data._headers = response.headers
-
-        return data
-
-    def retrieve(self, thread_id, run_id, step_id, **kwargs) -> RunStep:
+    def retrieve(
+        self,
+        thread_id,
+        run_id,
+        step_id,
+        include: Union[List[Any], NotGiven] = NOT_GIVEN,
+        **kwargs,
+    ) -> RunStep:
         if kwargs:
             response = (
                 self.openai_client.with_raw_response.beta.threads.runs.steps.retrieve(
                     thread_id=thread_id,
                     run_id=run_id,
                     step_id=step_id,
+                    include=include,
                     extra_body=kwargs,
                 )
             )
         else:
             response = (
                 self.openai_client.with_raw_response.beta.threads.runs.steps.retrieve(
-                    thread_id=thread_id, run_id=run_id, step_id=step_id
+                    thread_id=thread_id,
+                    run_id=run_id,
+                    step_id=step_id,
+                    include=include,
                 )
             )
         data = RunStep(**json.loads(response.text))
@@ -717,6 +771,33 @@ def retrieve(self, thread_id, run_id, step_id, **kwargs) -> RunStep:
 
         return data
 
+    def list(
+        self,
+        run_id,
+        *,
+        thread_id,
+        after: Union[str, NotGiven] = NOT_GIVEN,
+        before: Union[str, NotGiven] = NOT_GIVEN,
+        include: Union[List[Any], NotGiven] = NOT_GIVEN,
+        limit: Union[int, NotGiven] = NOT_GIVEN,
+        order: Union[Any, NotGiven] = NOT_GIVEN,
+        **kwargs,
+    ) -> RunStepList:
+        response = self.openai_client.with_raw_response.beta.threads.runs.steps.list(
+            thread_id=thread_id,
+            run_id=run_id,
+            after=after,
+            before=before,
+            include=include,
+            limit=limit,
+            order=order,
+            **kwargs,
+        )
+        data = RunStepList(**json.loads(response.text))
+        data._headers = response.headers
+
+        return data
+
 
 class AsyncThreads(AsyncAPIResource):
     def __init__(self, client: AsyncPortkey) -> None:
@@ -965,6 +1046,36 @@ async def create(
 
         return data
 
+    async def retrieve(self, thread_id, message_id, **kwargs) -> ThreadMessage:
+        if kwargs:
+            response = await self.openai_client.with_raw_response.beta.threads.messages.retrieve(  # noqa: E501
+                thread_id=thread_id, message_id=message_id, extra_body=kwargs
+            )
+        else:
+            response = await self.openai_client.with_raw_response.beta.threads.messages.retrieve(  # noqa: E501
+                thread_id=thread_id, message_id=message_id
+            )
+        data = ThreadMessage(**json.loads(response.text))
+        data._headers = response.headers
+        return data
+
+    async def update(
+        self,
+        thread_id,
+        message_id,
+        *,
+        metadata: Union[Optional[object], NotGiven] = NOT_GIVEN,
+        **kwargs,
+    ) -> ThreadMessage:
+        response = (
+            await self.openai_client.with_raw_response.beta.threads.messages.update(
+                thread_id=thread_id, message_id=message_id, metadata=metadata, **kwargs
+            )
+        )
+        data = ThreadMessage(**json.loads(response.text))
+        data._headers = response.headers
+        return data
+
     async def list(
         self,
         thread_id,
@@ -991,29 +1102,6 @@ async def list(
         data._headers = response.headers
         return data
 
-    async def retrieve(self, thread_id, message_id, **kwargs) -> ThreadMessage:
-        if kwargs:
-            response = await self.openai_client.with_raw_response.beta.threads.messages.retrieve(  # noqa: E501
-                thread_id=thread_id, message_id=message_id, extra_body=kwargs
-            )
-        else:
-            response = await self.openai_client.with_raw_response.beta.threads.messages.retrieve(  # noqa: E501
-                thread_id=thread_id, message_id=message_id
-            )
-        data = ThreadMessage(**json.loads(response.text))
-        data._headers = response.headers
-        return data
-
-    async def update(self, thread_id, message_id, **kwargs) -> ThreadMessage:
-        response = (
-            await self.openai_client.with_raw_response.beta.threads.messages.update(
-                thread_id=thread_id, message_id=message_id, **kwargs
-            )
-        )
-        data = ThreadMessage(**json.loads(response.text))
-        data._headers = response.headers
-        return data
-
     async def delete(
         self, message_id: str, *, thread_id: str, **kwargs
     ) -> ThreadMessageDeleted:
@@ -1104,15 +1192,6 @@ async def retrieve(self, thread_id, run_id, **kwargs) -> Run:
 
         return data
 
-    async def list(self, thread_id, **kwargs) -> RunList:
-        response = await self.openai_client.with_raw_response.beta.threads.runs.list(
-            thread_id=thread_id, **kwargs
-        )
-        data = RunList(**json.loads(response.text))
-        data._headers = response.headers
-
-        return data
-
     async def update(
         self,
         thread_id,
@@ -1128,23 +1207,25 @@ async def update(
 
         return data
 
-    async def submit_tool_outputs(
-        self, thread_id, tool_outputs, run_id, **kwargs
-    ) -> Run:
-        # fmt: off
-        response = await self.openai_client\
-            .with_raw_response\
-            .beta\
-            .threads\
-            .runs\
-            .submit_tool_outputs(
-                thread_id=thread_id,
-                run_id=run_id,
-                tool_outputs=tool_outputs,
-                **kwargs
+    async def list(
+        self,
+        thread_id,
+        *,
+        after: Union[str, NotGiven] = NOT_GIVEN,
+        before: Union[str, NotGiven] = NOT_GIVEN,
+        limit: Union[int, NotGiven] = NOT_GIVEN,
+        order: Union[NotGiven, Literal["asc", "desc"]] = NOT_GIVEN,
+        **kwargs,
+    ) -> RunList:
+        response = await self.openai_client.with_raw_response.beta.threads.runs.list(
+            thread_id=thread_id,
+            after=after,
+            before=before,
+            limit=limit,
+            order=order,
+            **kwargs,
         )
-        # fmt: on
-        data = Run(**json.loads(response.text))
+        data = RunList(**json.loads(response.text))
         data._headers = response.headers
 
         return data
@@ -1162,6 +1243,7 @@ async def create_and_poll(
         self,
         *,
         assistant_id: str,
+        include: Union[List[Any], NotGiven] = NOT_GIVEN,
         additional_instructions: Union[Optional[str], NotGiven] = NOT_GIVEN,
         additional_messages: Union[
             Optional[Iterable[run_create_params.AdditionalMessage]], NotGiven
@@ -1171,6 +1253,7 @@ async def create_and_poll(
         max_prompt_tokens: Union[Optional[int], NotGiven] = NOT_GIVEN,
         metadata: Union[Optional[object], NotGiven] = NOT_GIVEN,
         model: Union[str, None, NotGiven] = NOT_GIVEN,
+        parallel_tool_calls: Union[bool, NotGiven] = NOT_GIVEN,
         response_format: Union[
             Optional[AssistantResponseFormatOptionParam], NotGiven
         ] = NOT_GIVEN,
@@ -1189,6 +1272,7 @@ async def create_and_poll(
     ) -> Run:
         response = await self.openai_client.beta.threads.runs.create_and_poll(
             assistant_id=assistant_id,
+            include=include,
             additional_instructions=additional_instructions,
             additional_messages=additional_messages,
             instructions=instructions,
@@ -1196,6 +1280,7 @@ async def create_and_poll(
             max_prompt_tokens=max_prompt_tokens,
             metadata=metadata,
             model=model,
+            parallel_tool_calls=parallel_tool_calls,
             response_format=response_format,
             temperature=temperature,
             tool_choice=tool_choice,
@@ -1224,6 +1309,7 @@ async def create_and_stream(
         max_prompt_tokens: Union[Optional[int], NotGiven] = NOT_GIVEN,
         metadata: Union[Optional[object], NotGiven] = NOT_GIVEN,
         model: Union[str, None, NotGiven] = NOT_GIVEN,
+        parallel_tool_calls: Union[bool, NotGiven] = NOT_GIVEN,
         response_format: Union[
             Optional[AssistantResponseFormatOptionParam], NotGiven
         ] = NOT_GIVEN,
@@ -1254,6 +1340,7 @@ async def create_and_stream(
             max_prompt_tokens=max_prompt_tokens,
             metadata=metadata,
             model=model,
+            parallel_tool_calls=parallel_tool_calls,
             response_format=response_format,
             temperature=temperature,
             tool_choice=tool_choice,
@@ -1286,6 +1373,7 @@ async def stream(
         self,
         *,
         assistant_id: str,
+        include: Union[List[Any], NotGiven] = NOT_GIVEN,
         additional_instructions: Union[Optional[str], NotGiven] = NOT_GIVEN,
         additional_messages: Union[
             Optional[Iterable[run_create_params.AdditionalMessage]], NotGiven
@@ -1295,6 +1383,7 @@ async def stream(
         max_prompt_tokens: Union[Optional[int], NotGiven] = NOT_GIVEN,
         metadata: Union[Optional[object], NotGiven] = NOT_GIVEN,
         model: Union[str, None, NotGiven] = NOT_GIVEN,
+        parallel_tool_calls: Union[bool, NotGiven] = NOT_GIVEN,
         response_format: Union[
             Optional[AssistantResponseFormatOptionParam], NotGiven
         ] = NOT_GIVEN,
@@ -1318,6 +1407,7 @@ async def stream(
     ):
         response = await self.openai_client.beta.threads.runs.stream(
             assistant_id=assistant_id,
+            include=include,
             additional_instructions=additional_instructions,
             additional_messages=additional_messages,
             instructions=instructions,
@@ -1325,6 +1415,7 @@ async def stream(
             max_prompt_tokens=max_prompt_tokens,
             metadata=metadata,
             model=model,
+            parallel_tool_calls=parallel_tool_calls,
             response_format=response_format,
             temperature=temperature,
             tool_choice=tool_choice,
@@ -1338,6 +1429,34 @@ async def stream(
         data = response
         return data
 
+    async def submit_tool_outputs(
+        self,
+        run_id,
+        *,
+        thread_id,
+        tool_outputs,
+        stream: Union[bool, NotGiven] = NOT_GIVEN,
+        **kwargs,
+    ) -> Run:
+        # fmt: off
+        response = await self.openai_client\
+            .with_raw_response\
+            .beta\
+            .threads\
+            .runs\
+            .submit_tool_outputs(
+                thread_id=thread_id,
+                run_id=run_id,
+                tool_outputs=tool_outputs,
+                stream=stream,
+                **kwargs
+        )
+        # fmt: on
+        data = Run(**json.loads(response.text))
+        data._headers = response.headers
+
+        return data
+
     async def submit_tool_outputs_and_poll(
         self,
         *,
@@ -1389,27 +1508,56 @@ def __init__(self, client: AsyncPortkey) -> None:
         super().__init__(client)
         self.openai_client = client.openai_client
 
-    async def list(self, thread_id, run_id, **kwargs) -> RunStepList:
-        response = (
-            await self.openai_client.with_raw_response.beta.threads.runs.steps.list(
-                thread_id=thread_id, run_id=run_id, **kwargs
-            )
-        )
-        data = RunStepList(**json.loads(response.text))
-        data._headers = response.headers
-
-        return data
-
-    async def retrieve(self, thread_id, run_id, step_id, **kwargs) -> RunStep:
+    async def retrieve(
+        self,
+        thread_id,
+        run_id,
+        step_id,
+        include: Union[List[Any], NotGiven] = NOT_GIVEN,
+        **kwargs,
+    ) -> RunStep:
         if kwargs:
             response = await self.openai_client.with_raw_response.beta.threads.runs.steps.retrieve(  # noqa: E501
-                thread_id=thread_id, run_id=run_id, step_id=step_id, extra_body=kwargs
+                thread_id=thread_id,
+                run_id=run_id,
+                step_id=step_id,
+                include=include,
+                extra_body=kwargs,
             )
         else:
             response = await self.openai_client.with_raw_response.beta.threads.runs.steps.retrieve(  # noqa: E501
-                thread_id=thread_id, run_id=run_id, step_id=step_id
+                thread_id=thread_id, run_id=run_id, step_id=step_id, include=include
             )
         data = RunStep(**json.loads(response.text))
         data._headers = response.headers
 
         return data
+
+    async def list(
+        self,
+        run_id,
+        *,
+        thread_id,
+        after: Union[str, NotGiven] = NOT_GIVEN,
+        before: Union[str, NotGiven] = NOT_GIVEN,
+        include: Union[List[Any], NotGiven] = NOT_GIVEN,
+        limit: Union[int, NotGiven] = NOT_GIVEN,
+        order: Union[Any, NotGiven] = NOT_GIVEN,
+        **kwargs,
+    ) -> RunStepList:
+        response = (
+            await self.openai_client.with_raw_response.beta.threads.runs.steps.list(
+                thread_id=thread_id,
+                run_id=run_id,
+                after=after,
+                before=before,
+                include=include,
+                limit=limit,
+                order=order,
+                **kwargs,
+            )
+        )
+        data = RunStepList(**json.loads(response.text))
+        data._headers = response.headers
+
+        return data
diff --git a/portkey_ai/api_resources/apis/uploads.py b/portkey_ai/api_resources/apis/uploads.py
index 49364490..64343590 100644
--- a/portkey_ai/api_resources/apis/uploads.py
+++ b/portkey_ai/api_resources/apis/uploads.py
@@ -1,5 +1,7 @@
 import json
+import os
 from typing import Any, List, Union
+import typing
 from portkey_ai.api_resources.apis.api_resource import APIResource, AsyncAPIResource
 from portkey_ai.api_resources.client import AsyncPortkey, Portkey
 from portkey_ai.api_resources.types.upload_types import Upload, UploadPart
@@ -12,6 +14,29 @@ def __init__(self, client: Portkey) -> None:
         self.openai_client = client.openai_client
         self.parts = Parts(client)
 
+    @typing.no_type_check
+    def upload_file_chunked(
+        self,
+        *,
+        file: Union[os.PathLike[str], bytes],
+        mime_type: str,
+        purpose: Any,
+        filename: Union[str, None] = None,
+        bytes: Union[int, None] = None,
+        part_size: Union[int, None] = None,
+        md5: Union[str, NotGiven] = NOT_GIVEN,
+    ) -> Any:
+        response = self.openai_client.uploads.upload_file_chunked(
+            file=file,
+            mime_type=mime_type,
+            purpose=purpose,
+            filename=filename,
+            bytes=bytes,
+            part_size=part_size,
+            md5=md5,
+        )
+        return response
+
     def create(
         self, *, bytes: int, filename: str, mime_type: str, purpose: Any, **kwargs
     ) -> Upload:
@@ -45,7 +70,7 @@ def complete(
         *,
         part_ids: List[str],
         md5: Union[str, NotGiven] = NOT_GIVEN,
-        **kwargs
+        **kwargs,
     ) -> Upload:
         extra_headers = kwargs.pop("extra_headers", {})
         response = self.openai_client.with_raw_response.uploads.complete(
@@ -86,6 +111,29 @@ def __init__(self, client: AsyncPortkey) -> None:
         self.openai_client = client.openai_client
         self.parts = AsyncParts(client)
 
+    @typing.no_type_check
+    async def upload_file_chunked(
+        self,
+        *,
+        file: Union[os.PathLike[str], bytes],
+        mime_type: str,
+        purpose: Any,
+        filename: Union[str, None] = None,
+        bytes: Union[int, None] = None,
+        part_size: Union[int, None] = None,
+        md5: Union[str, NotGiven] = NOT_GIVEN,
+    ) -> Any:
+        response = await self.openai_client.uploads.upload_file_chunked(
+            file=file,
+            mime_type=mime_type,
+            purpose=purpose,
+            filename=filename,
+            bytes=bytes,
+            part_size=part_size,
+            md5=md5,
+        )
+        return response
+
     async def create(
         self, *, bytes: int, filename: str, mime_type: str, purpose: Any, **kwargs
     ) -> Upload:
@@ -119,7 +167,7 @@ async def complete(
         *,
         part_ids: List[str],
         md5: Union[str, NotGiven] = NOT_GIVEN,
-        **kwargs
+        **kwargs,
     ) -> Upload:
         extra_headers = kwargs.pop("extra_headers", {})
         response = await self.openai_client.with_raw_response.uploads.complete(
diff --git a/portkey_ai/api_resources/apis/vector_stores.py b/portkey_ai/api_resources/apis/vector_stores.py
index 9258822f..cee336d5 100644
--- a/portkey_ai/api_resources/apis/vector_stores.py
+++ b/portkey_ai/api_resources/apis/vector_stores.py
@@ -97,12 +97,14 @@ def list(
         self,
         *,
         after: Union[str, NotGiven] = NOT_GIVEN,
+        before: Union[str, NotGiven] = NOT_GIVEN,
         limit: Union[int, NotGiven] = NOT_GIVEN,
         order: Union[str, NotGiven] = NOT_GIVEN,
         **kwargs,
     ) -> VectorStoreList:
         response = self.openai_client.with_raw_response.beta.vector_stores.list(
             after=after,
+            before=before,
             limit=limit,
             order=order,
             **kwargs,
@@ -398,7 +400,7 @@ def list_files(
             order=order,
             **kwargs,
         )
-        data = VectorStoreFileBatch(**json.loads(response.text))
+        data = VectorStoreFileList(**json.loads(response.text))
         data._headers = response.headers
 
         return data
@@ -527,12 +529,14 @@ async def list(
         self,
         *,
         after: Union[str, NotGiven] = NOT_GIVEN,
+        before: Union[str, NotGiven] = NOT_GIVEN,
         limit: Union[int, NotGiven] = NOT_GIVEN,
         order: Union[str, NotGiven] = NOT_GIVEN,
         **kwargs,
     ) -> VectorStoreList:
         response = await self.openai_client.with_raw_response.beta.vector_stores.list(
             after=after,
+            before=before,
             limit=limit,
             order=order,
             **kwargs,
diff --git a/portkey_ai/api_resources/base_client.py b/portkey_ai/api_resources/base_client.py
index b5d918cd..dd70adc9 100644
--- a/portkey_ai/api_resources/base_client.py
+++ b/portkey_ai/api_resources/base_client.py
@@ -53,6 +53,7 @@ def __init__(
         base_url: Optional[str] = None,
         api_key: Optional[str] = None,
         virtual_key: Optional[str] = None,
+        websocket_base_url: Optional[Union[str, httpx.URL]] = None,
         config: Optional[Union[Mapping, str]] = None,
         provider: Optional[str] = None,
         trace_id: Optional[str] = None,
@@ -87,6 +88,7 @@ def __init__(
         self.base_url = set_base_url(base_url, api_key)
         self.api_key = default_api_key(self.base_url, api_key)
         self.virtual_key = virtual_key
+        self.websocket_base_url = websocket_base_url
         self.config = config
         self.provider = provider
         self.trace_id = trace_id
@@ -697,6 +699,7 @@ def __init__(
         base_url: Optional[str] = None,
         api_key: Optional[str] = None,
         virtual_key: Optional[str] = None,
+        websocket_base_url: Optional[Union[str, httpx.URL]] = None,
         config: Optional[Union[Mapping, str]] = None,
         provider: Optional[str] = None,
         trace_id: Optional[str] = None,
@@ -731,6 +734,7 @@ def __init__(
         self.base_url = set_base_url(base_url, api_key)
         self.api_key = default_api_key(self.base_url, api_key)
         self.virtual_key = virtual_key
+        self.websocket_base_url = websocket_base_url
         self.config = config
         self.provider = provider
         self.trace_id = trace_id
diff --git a/portkey_ai/api_resources/client.py b/portkey_ai/api_resources/client.py
index 0d0951b9..01c6bc89 100644
--- a/portkey_ai/api_resources/client.py
+++ b/portkey_ai/api_resources/client.py
@@ -38,12 +38,14 @@ class beta:
         threads: apis.Threads
         vector_stores: apis.VectorStores
         chat: apis.BetaChat
+        realtime: apis.BetaRealtime
 
         def __init__(self, client: Portkey) -> None:
             self.assistants = apis.Assistants(client)
             self.threads = apis.Threads(client)
             self.vector_stores = apis.VectorStores(client)
             self.chat = apis.BetaChat(client)
+            self.realtime = apis.BetaRealtime(client)
 
     def __init__(
         self,
@@ -51,6 +53,7 @@ def __init__(
         api_key: Optional[str] = None,
         base_url: Optional[str] = None,
         virtual_key: Optional[str] = None,
+        websocket_base_url: Optional[Union[str, httpx.URL]] = None,
         config: Optional[Union[Mapping, str]] = None,
         provider: Optional[str] = None,
         trace_id: Optional[str] = None,
@@ -86,6 +89,7 @@ def __init__(
             api_key=api_key,
             base_url=base_url,
             virtual_key=virtual_key,
+            websocket_base_url=websocket_base_url,
             config=config,
             provider=provider,
             trace_id=trace_id,
@@ -124,6 +128,7 @@ def __init__(
             default_headers=self.allHeaders,
             http_client=http_client,
             max_retries=0,
+            websocket_base_url=self.websocket_base_url,
         )
 
         self.completions = apis.Completion(self)
@@ -153,6 +158,7 @@ def copy(
         api_key: Optional[str] = None,
         base_url: Optional[str] = None,
         virtual_key: Optional[str] = None,
+        websocket_base_url: Optional[Union[str, httpx.URL]] = None,
         config: Optional[Union[Mapping, str]] = None,
         provider: Optional[str] = None,
         trace_id: Optional[str] = None,
@@ -188,6 +194,7 @@ def copy(
             api_key=api_key or self.api_key,
             base_url=base_url or self.base_url,
             virtual_key=virtual_key or self.virtual_key,
+            websocket_base_url=websocket_base_url or self.websocket_base_url,
             config=config or self.config,
             provider=provider or self.provider,
             trace_id=trace_id or self.trace_id,
@@ -255,12 +262,14 @@ class beta:
         threads: apis.AsyncThreads
         vector_stores: apis.AsyncVectorStores
         chat: apis.AsyncBetaChat
+        realtime: apis.AsyncBetaRealtime
 
         def __init__(self, client: AsyncPortkey) -> None:
             self.assistants = apis.AsyncAssistants(client)
             self.threads = apis.AsyncThreads(client)
             self.vector_stores = apis.AsyncVectorStores(client)
             self.chat = apis.AsyncBetaChat(client)
+            self.realtime = apis.AsyncBetaRealtime(client)
 
     def __init__(
         self,
@@ -268,6 +277,7 @@ def __init__(
         api_key: Optional[str] = None,
         base_url: Optional[str] = None,
         virtual_key: Optional[str] = None,
+        websocket_base_url: Optional[Union[str, httpx.URL]] = None,
         config: Optional[Union[Mapping, str]] = None,
         provider: Optional[str] = None,
         trace_id: Optional[str] = None,
@@ -303,6 +313,7 @@ def __init__(
             api_key=api_key,
             base_url=base_url,
             virtual_key=virtual_key,
+            websocket_base_url=websocket_base_url,
             config=config,
             provider=provider,
             trace_id=trace_id,
@@ -341,6 +352,7 @@ def __init__(
             default_headers=self.allHeaders,
             http_client=http_client,
             max_retries=0,
+            websocket_base_url=self.websocket_base_url,
         )
 
         self.completions = apis.AsyncCompletion(self)
@@ -370,6 +382,7 @@ def copy(
         api_key: Optional[str] = None,
         base_url: Optional[str] = None,
         virtual_key: Optional[str] = None,
+        websocket_base_url: Optional[Union[str, httpx.URL]] = None,
         config: Optional[Union[Mapping, str]] = None,
         provider: Optional[str] = None,
         trace_id: Optional[str] = None,
@@ -405,6 +418,7 @@ def copy(
             api_key=api_key or self.api_key,
             base_url=base_url or self.base_url,
             virtual_key=virtual_key or self.virtual_key,
+            websocket_base_url=websocket_base_url or self.websocket_base_url,
             config=config or self.config,
             provider=provider or self.provider,
             trace_id=trace_id or self.trace_id,
diff --git a/portkey_ai/api_resources/types/assistant_type.py b/portkey_ai/api_resources/types/assistant_type.py
index fa17caa2..9197663b 100644
--- a/portkey_ai/api_resources/types/assistant_type.py
+++ b/portkey_ai/api_resources/types/assistant_type.py
@@ -17,11 +17,20 @@
     "FunctionParameters",
     "FileSearch",
     "FileSearchTool",
+    "ToolResources",
+    "ToolResourcesCodeInterpreter",
+    "ToolResourcesFileSearch",
 ]
 
 
+class FileSearchRankingOptions(BaseModel, extra="allow"):
+    score_threshold: Optional[float] = None
+    ranker: Optional[str] = None
+
+
 class FileSearch(BaseModel, extra="allow"):
     max_num_results: Optional[int] = None
+    ranking_options: Optional[FileSearchRankingOptions] = None
 
 
 class FileSearchTool(BaseModel, extra="allow"):
@@ -48,13 +57,26 @@ class ToolRetrieval(BaseModel, extra="allow"):
 
 
 class ToolFunction(BaseModel, extra="allow"):
-    type: Optional[str] = None
     function: Optional[FunctionDefinition] = None
+    type: Optional[str] = None
 
 
 AssistantTool = Union[ToolCodeInterpreter, ToolRetrieval, ToolFunction, FileSearchTool]
 
 
+class ToolResourcesCodeInterpreter(BaseModel, extra="allow"):
+    file_ids: Optional[List[str]] = None
+
+
+class ToolResourcesFileSearch(BaseModel, extra="allow"):
+    vector_store_ids: Optional[List[str]] = None
+
+
+class ToolResources(BaseModel, extra="allow"):
+    code_interpreter: Optional[ToolResourcesCodeInterpreter] = None
+    file_search: Optional[ToolResourcesFileSearch] = None
+
+
 class Assistant(BaseModel, extra="allow"):
     id: Optional[str] = None
     created_at: Optional[int] = None
@@ -67,6 +89,9 @@ class Assistant(BaseModel, extra="allow"):
     object: Optional[str] = None
     tools: Optional[List[AssistantTool]] = None
     response_format: Optional[Any] = None
+    temperature: Optional[float] = None
+    tool_resources: Optional[ToolResources] = None
+    top_p: Optional[float] = None
     _headers: Optional[httpx.Headers] = PrivateAttr()
 
     def __str__(self):
diff --git a/portkey_ai/api_resources/types/audio_types.py b/portkey_ai/api_resources/types/audio_types.py
index 7ae35605..0a282d7d 100644
--- a/portkey_ai/api_resources/types/audio_types.py
+++ b/portkey_ai/api_resources/types/audio_types.py
@@ -1,5 +1,5 @@
 import json
-from typing import Dict, Optional
+from typing import Dict, List, Optional
 import httpx
 from .utils import parse_headers
 from typing import Any
@@ -8,6 +8,47 @@
 __all__ = ["Transcription", "Translation"]
 
 
+class TranscriptionSegment(BaseModel, extra="allow"):
+    id: Optional[int] = None
+    avg_logprob: Optional[float] = None
+    compression_ratio: Optional[float] = None
+    end: Optional[float] = None
+    no_speech_prob: Optional[float] = None
+    seek: Optional[int] = None
+    start: Optional[float] = None
+    temperature: Optional[float] = None
+    text: Optional[str] = None
+    tokens: Optional[List[int]] = None
+
+
+class TranscriptionWord(BaseModel, extra="allow"):
+    start: Optional[float] = None
+    end: Optional[float] = None
+    word: Optional[str] = None
+
+
+class TranscriptionVerbose(BaseModel, extra="allow"):
+    duration: Optional[str] = None
+    language: Optional[str] = None
+    text: Optional[str] = None
+    segments: Optional[List[TranscriptionSegment]] = None
+    words: Optional[List[TranscriptionWord]] = None
+    _headers: Optional[httpx.Headers] = PrivateAttr()
+
+    def __str__(self):
+        del self._headers
+        return json.dumps(self.dict(), indent=4)
+
+    def __getitem__(self, key):
+        return getattr(self, key, None)
+
+    def get(self, key: str, default: Optional[Any] = None):
+        return getattr(self, key, None) or default
+
+    def get_headers(self) -> Optional[Dict[str, str]]:
+        return parse_headers(self._headers)
+
+
 class Transcription(BaseModel, extra="allow"):
     text: Optional[str] = None
     _headers: Optional[httpx.Headers] = PrivateAttr()
@@ -26,6 +67,27 @@ def get_headers(self) -> Optional[Dict[str, str]]:
         return parse_headers(self._headers)
 
 
+class TranslationVerbose(BaseModel, extra="allow"):
+    duration: Optional[str] = None
+    language: Optional[str] = None
+    text: Optional[str] = None
+    segments: Optional[List[TranscriptionSegment]] = None
+    _headers: Optional[httpx.Headers] = PrivateAttr()
+
+    def __str__(self):
+        del self._headers
+        return json.dumps(self.dict(), indent=4)
+
+    def __getitem__(self, key):
+        return getattr(self, key, None)
+
+    def get(self, key: str, default: Optional[Any] = None):
+        return getattr(self, key, None) or default
+
+    def get_headers(self) -> Optional[Dict[str, str]]:
+        return parse_headers(self._headers)
+
+
 class Translation(BaseModel, extra="allow"):
     text: Optional[str] = None
     _headers: Optional[httpx.Headers] = PrivateAttr()
diff --git a/portkey_ai/api_resources/types/beta_realtime.py b/portkey_ai/api_resources/types/beta_realtime.py
new file mode 100644
index 00000000..7eed3018
--- /dev/null
+++ b/portkey_ai/api_resources/types/beta_realtime.py
@@ -0,0 +1,66 @@
+import json
+from typing import Any, Dict, List, Optional, Union
+import httpx
+from pydantic import BaseModel, PrivateAttr
+from portkey_ai.api_resources.types.utils import parse_headers
+
+__all__ = [
+    "SessionCreateResponse",
+    "ClientSecret",
+    "InputAudioTranscription",
+    "Tool",
+    "TurnDetection",
+]
+
+
+class ClientSecret(BaseModel, extra="allow"):
+    expires_at: Optional[int] = None
+    value: Optional[str] = None
+
+
+class InputAudioTranscription(BaseModel, extra="allow"):
+    model: Optional[str] = None
+
+
+class Tool(BaseModel, extra="allow"):
+    description: Optional[str] = None
+    name: Optional[str] = None
+    parameters: Optional[object] = None
+    type: Optional[str] = None
+
+
+class TurnDetection(BaseModel, extra="allow"):
+    prefix_padding_ms: Optional[int] = None
+    silence_duration_ms: Optional[int] = None
+    threshold: Optional[float] = None
+    type: Optional[str] = None
+
+
+class SessionCreateResponse(BaseModel, extra="allow"):
+    client_secret: Optional[ClientSecret] = None
+    input_audio_format: Optional[str] = None
+    input_audio_transcription: Optional[InputAudioTranscription] = None
+    instructions: Optional[str] = None
+    max_response_output_tokens: Union[int, str, None] = None
+    modalities: Optional[List[str]] = None
+    output_audio_format: Optional[str] = None
+    temperature: Optional[float] = None
+    tool_choice: Optional[str] = None
+    tools: Optional[List[Tool]] = None
+    turn_detection: Optional[TurnDetection] = None
+    voice: Optional[str] = None
+
+    _headers: Optional[httpx.Headers] = PrivateAttr()
+
+    def __str__(self):
+        del self._headers
+        return json.dumps(self.dict(), indent=4)
+
+    def __getitem__(self, key):
+        return getattr(self, key, None)
+
+    def get(self, key: str, default: Optional[Any] = None):
+        return getattr(self, key, None) or default
+
+    def get_headers(self) -> Optional[Dict[str, str]]:
+        return parse_headers(self._headers)
diff --git a/portkey_ai/api_resources/types/chat_complete_type.py b/portkey_ai/api_resources/types/chat_complete_type.py
index 66be5e1c..ba5456c7 100644
--- a/portkey_ai/api_resources/types/chat_complete_type.py
+++ b/portkey_ai/api_resources/types/chat_complete_type.py
@@ -41,10 +41,24 @@ class ChoiceLogprobs(BaseModel, extra="allow"):
     refusal: Optional[List[ChatCompletionTokenLogprob]] = None
 
 
+class CompletionTokensDetails(BaseModel, extra="allow"):
+    accepted_prediction_tokens: Optional[int] = None
+    audio_tokens: Optional[int] = None
+    reasoning_tokens: Optional[int] = None
+    rejected_prediction_tokens: Optional[int] = None
+
+
+class PromptTokensDetails(BaseModel, extra="allow"):
+    audio_tokens: Optional[int] = None
+    cached_tokens: Optional[int] = None
+
+
 class Usage(BaseModel, extra="allow"):
     prompt_tokens: Optional[int] = None
     completion_tokens: Optional[int] = None
     total_tokens: Optional[int] = None
+    completion_tokens_details: Optional[CompletionTokensDetails] = None
+    prompt_tokens_details: Optional[PromptTokensDetails] = None
 
 
 class DeltaToolCallFunction(BaseModel, extra="allow"):
@@ -93,12 +107,20 @@ class ChatCompletionMessageToolCall(BaseModel, extra="allow"):
     type: Optional[str] = None
 
 
+class ChatCompletionAudio(BaseModel, extra="allow"):
+    id: Optional[str] = None
+    data: Optional[str] = None
+    expires_at: Optional[int] = None
+    transcript: Optional[str] = None
+
+
 class ChatCompletionMessage(BaseModel, extra="allow"):
     content: Optional[Union[str, Iterable[Any]]] = None
     role: Optional[str] = None
     function_call: Optional[FunctionCall] = None
     tool_calls: Optional[List[ChatCompletionMessageToolCall]] = None
     refusal: Optional[str] = None
+    audio: Optional[ChatCompletionAudio] = None
 
 
 class Choice(BaseModel, extra="allow"):
@@ -140,6 +162,8 @@ class ChatCompletionChunk(BaseModel, extra="allow"):
     model: Optional[str] = None
     choices: Optional[List[StreamChoice]] = None
     service_tier: Optional[str] = None
+    system_fingerprint: Optional[str] = None
+    usage: Optional[Usage] = None
 
     def __str__(self):
         return json.dumps(self.dict(), indent=4)
diff --git a/portkey_ai/api_resources/types/fine_tuning_type.py b/portkey_ai/api_resources/types/fine_tuning_type.py
index 71a166d0..34169b57 100644
--- a/portkey_ai/api_resources/types/fine_tuning_type.py
+++ b/portkey_ai/api_resources/types/fine_tuning_type.py
@@ -17,6 +17,11 @@
     "FineTuningJobCheckpointList",
     "FineTuningJobWandbIntegration",
     "FineTuningJobWandbIntegrationObject",
+    "MethodDpoHyperparameters",
+    "MethodSupervisedHyperparameters",
+    "MethodDpo",
+    "MethodSupervised",
+    "Method",
 ]
 
 
@@ -27,6 +32,8 @@ class Error(BaseModel, extra="allow"):
 
 
 class Hyperparameters(BaseModel, extra="allow"):
+    batch_size: Optional[Union[str, int]] = None
+    learning_rate_multiplier: Optional[Union[str, float]] = None
     n_epochs: Optional[Union[str, int]] = None
 
 
@@ -42,6 +49,33 @@ class FineTuningJobWandbIntegrationObject(BaseModel, extra="allow"):
     wandb: Optional[FineTuningJobWandbIntegration] = None
 
 
+class MethodDpoHyperparameters(BaseModel, extra="allow"):
+    batch_size: Optional[Union[str, int]] = None
+    beta: Optional[Union[str, float]] = None
+    learning_rate_multiplier: Optional[Union[str, float]] = None
+    n_epochs: Optional[Union[str, int]] = None
+
+
+class MethodSupervisedHyperparameters(BaseModel, extra="allow"):
+    batch_size: Optional[Union[str, int]] = None
+    learning_rate_multiplier: Optional[Union[str, float]] = None
+    n_epochs: Optional[Union[str, int]] = None
+
+
+class MethodDpo(BaseModel, extra="allow"):
+    hyperparameters: Optional[MethodDpoHyperparameters] = None
+
+
+class MethodSupervised(BaseModel, extra="allow"):
+    hyperparameters: Optional[MethodSupervisedHyperparameters] = None
+
+
+class Method(BaseModel, extra="allow"):
+    dpo: Optional[MethodDpo] = None
+    supervised: Optional[MethodSupervised] = None
+    type: Optional[str] = None
+
+
 class FineTuningJob(BaseModel, extra="allow"):
     id: Optional[str] = None
     created_at: Optional[int] = None
@@ -60,6 +94,7 @@ class FineTuningJob(BaseModel, extra="allow"):
     validation_file: Optional[str] = None
     estimated_finish: Optional[int] = None
     integrations: Optional[List[FineTuningJobWandbIntegrationObject]] = None
+    method: Optional[Method] = None
     _headers: Optional[httpx.Headers] = PrivateAttr()
 
     def __str__(self):
@@ -101,6 +136,8 @@ class FineTuningJobEvent(BaseModel, extra="allow"):
     level: Optional[str] = None
     message: Optional[str] = None
     object: Optional[str] = None
+    data: Optional[Any] = None
+    type: Optional[str] = None
     _headers: Optional[httpx.Headers] = PrivateAttr()
 
     def __str__(self):
diff --git a/portkey_ai/api_resources/types/moderations_type.py b/portkey_ai/api_resources/types/moderations_type.py
index 9d9efcb5..75b27bd2 100644
--- a/portkey_ai/api_resources/types/moderations_type.py
+++ b/portkey_ai/api_resources/types/moderations_type.py
@@ -4,10 +4,70 @@
 from .utils import parse_headers
 from typing import List, Any
 from pydantic import BaseModel, PrivateAttr
-from ..._vendor.openai.types.moderation import Moderation
 
 
-__all__ = ["ModerationCreateResponse"]
+__all__ = [
+    "ModerationCreateResponse",
+    "Categories",
+    "CategoryAppliedInputTypes",
+    "CategoryScores",
+    "Moderation",
+]
+
+
+class Categories(BaseModel, extra="allow"):
+    harassment: Optional[bool] = None
+    harassment_threatening: Optional[bool] = None
+    hate: Optional[bool] = None
+    hate_threatening: Optional[bool] = None
+    illicit: Optional[bool] = None
+    illicit_violent: Optional[bool] = None
+    self_harm: Optional[bool] = None
+    self_harm_instructions: Optional[bool] = None
+    self_harm_intent: Optional[bool] = None
+    sexual: Optional[bool] = None
+    sexual_minors: Optional[bool] = None
+    violence: Optional[bool] = None
+    violence_graphic: Optional[bool] = None
+
+
+class CategoryAppliedInputTypes(BaseModel, extra="allow"):
+    harassment: Optional[List[str]] = None
+    harassment_threatening: Optional[List[str]] = None
+    hate: Optional[List[str]] = None
+    hate_threatening: Optional[List[str]] = None
+    illicit: Optional[List[str]] = None
+    illicit_violent: Optional[List[str]] = None
+    self_harm: Optional[List[str]] = None
+    self_harm_instructions: Optional[List[str]] = None
+    self_harm_intent: Optional[List[str]] = None
+    sexual: Optional[List[str]] = None
+    sexual_minors: Optional[List[str]] = None
+    violence: Optional[List[str]] = None
+    violence_graphic: Optional[List[str]] = None
+
+
+class CategoryScores(BaseModel, extra="allow"):
+    harassment: Optional[float] = None
+    harassment_threatening: Optional[float] = None
+    hate: Optional[float] = None
+    hate_threatening: Optional[float] = None
+    illicit: Optional[float] = None
+    illicit_violent: Optional[float] = None
+    self_harm: Optional[float] = None
+    self_harm_instructions: Optional[float] = None
+    self_harm_intent: Optional[float] = None
+    sexual: Optional[float] = None
+    sexual_minors: Optional[float] = None
+    violence: Optional[float] = None
+    violence_graphic: Optional[float] = None
+
+
+class Moderation(BaseModel, extra="allow"):
+    categories: Optional[Categories] = None
+    category_applied_input_types: Optional[CategoryAppliedInputTypes] = None
+    category_scores: Optional[CategoryScores] = None
+    flagged: Optional[bool] = None
 
 
 class ModerationCreateResponse(BaseModel, extra="allow"):
diff --git a/portkey_ai/api_resources/types/thread_message_type.py b/portkey_ai/api_resources/types/thread_message_type.py
index f7ceb0a9..b7ea7577 100644
--- a/portkey_ai/api_resources/types/thread_message_type.py
+++ b/portkey_ai/api_resources/types/thread_message_type.py
@@ -83,7 +83,6 @@ class Text(BaseModel, extra="allow"):
 
 class RefusalContentBlock(BaseModel, extra="allow"):
     refusal: Optional[str] = None
-
     type: Optional[str] = None
 
 
@@ -104,6 +103,7 @@ class MessageContentText(BaseModel, extra="allow"):
 
 class ImageFile(BaseModel, extra="allow"):
     file_id: Optional[str] = None
+    detail: Optional[str] = None
 
 
 class MessageContentImageFile(BaseModel, extra="allow"):
diff --git a/portkey_ai/api_resources/types/thread_run_type.py b/portkey_ai/api_resources/types/thread_run_type.py
index 5ffe8693..9d98a35a 100644
--- a/portkey_ai/api_resources/types/thread_run_type.py
+++ b/portkey_ai/api_resources/types/thread_run_type.py
@@ -1,5 +1,5 @@
 import json
-from typing import Dict, Literal, Optional, Union
+from typing import Any, Dict, Literal, Optional, Union
 import httpx
 
 from portkey_ai.api_resources.types.assistant_type import AssistantTool
@@ -42,6 +42,11 @@
     "AssistantToolChoiceFunction",
     "AssistantToolChoiceOption",
     "TruncationStrategy",
+    "FileSearchRankingOptions",
+    "FileSearchResultContent",
+    "FileSearchResult",
+    "FileSearch",
+    "FileSearchToolCall",
 ]
 
 
@@ -91,13 +96,35 @@ class CodeToolCall(BaseModel, extra="allow"):
     type: Optional[str] = None
 
 
+class FileSearchRankingOptions(BaseModel, extra="allow"):
+    ranker: Optional[str] = None
+    score_threshold: Optional[float] = None
+
+
+class FileSearchResultContent(BaseModel, extra="allow"):
+    text: Optional[str] = None
+    type: Optional[str] = None
+
+
+class FileSearchResult(BaseModel, extra="allow"):
+    file_id: Optional[str] = None
+    file_name: Optional[str] = None
+    score: Optional[float] = None
+    content: Optional[List[FileSearchResultContent]] = None
+
+
+class FileSearch(BaseModel, extra="allow"):
+    ranking_options: Optional[FileSearchRankingOptions] = None
+    results: Optional[List[FileSearchResult]] = None
+
+
 class FileSearchToolCall(BaseModel, extra="allow"):
     id: Optional[str] = None
-    file_search: Optional[object] = None
+    file_search: Optional[FileSearch] = None
     type: Optional[str] = None
 
 
-ToolCall = Union[CodeToolCall, RetrievalToolCall, FunctionToolCall, FileSearchToolCall]
+ToolCall = Union[CodeToolCall, FileSearchToolCall, RetrievalToolCall, FunctionToolCall]
 
 
 class ToolCallsStepDetails(BaseModel, extra="allow"):
@@ -226,6 +253,7 @@ class Run(BaseModel, extra="allow"):
     truncation_strategy: Optional[TruncationStrategy] = None
     temperature: Optional[float] = None
     top_p: Optional[float] = None
+    response_format: Optional[Any] = None
     _headers: Optional[httpx.Headers] = PrivateAttr()
 
     def __str__(self):
diff --git a/portkey_ai/api_resources/types/thread_type.py b/portkey_ai/api_resources/types/thread_type.py
index f980d4a2..4c4992fb 100644
--- a/portkey_ai/api_resources/types/thread_type.py
+++ b/portkey_ai/api_resources/types/thread_type.py
@@ -24,7 +24,6 @@ class ToolResourcesFileSearch(BaseModel, extra="allow"):
 
 class ToolResources(BaseModel, extra="allow"):
     code_interpreter: Optional[ToolResourcesCodeInterpreter] = None
-
     file_search: Optional[ToolResourcesFileSearch] = None
 
 
diff --git a/portkey_ai/api_resources/types/vector_stores_type.py b/portkey_ai/api_resources/types/vector_stores_type.py
index 3ff4a406..9ebc9f71 100644
--- a/portkey_ai/api_resources/types/vector_stores_type.py
+++ b/portkey_ai/api_resources/types/vector_stores_type.py
@@ -4,10 +4,6 @@
 import httpx
 
 from portkey_ai._vendor.openai._utils._transform import PropertyInfo
-from portkey_ai._vendor.openai.types.beta.vector_stores.vector_store_file import (
-    ChunkingStrategyOther,
-    ChunkingStrategyStatic,
-)
 from .utils import parse_headers
 from pydantic import BaseModel, PrivateAttr
 
@@ -92,8 +88,22 @@ class LastError(BaseModel, extra="allow"):
     message: Optional[str] = None
 
 
+class StaticFileChunkingStrategy(BaseModel, extra="allow"):
+    chunk_overlap_tokens: Optional[int] = None
+    max_chunk_size_tokens: Optional[int] = None
+
+
+class StaticFileChunkingStrategyObject(BaseModel, extra="allow"):
+    static: StaticFileChunkingStrategy
+    type: Optional[str] = None
+
+
+class OtherFileChunkingStrategyObject(BaseModel, extra="allow"):
+    type: Optional[str] = None
+
+
 ChunkingStrategy: TypeAlias = Annotated[
-    Union[ChunkingStrategyStatic, ChunkingStrategyOther],
+    Union[StaticFileChunkingStrategyObject, OtherFileChunkingStrategyObject],
     PropertyInfo(discriminator="type"),
 ]
 
diff --git a/vendorize.toml b/vendorize.toml
index 772c4675..e06951fb 100644
--- a/vendorize.toml
+++ b/vendorize.toml
@@ -1,4 +1,4 @@
 target = "portkey_ai/_vendor"
 packages = [
-    "openai==1.40.1"
+    "openai==1.58.1"
 ]
\ No newline at end of file