docs: add Wallaby provider page (wallaby/kimi-k3)

docs/providers/wallaby.md

@@ -0,0 +1,124 @@
+# Wallaby
+
+## Overview
+
+| Property | Details |
+|-------|-------|
+| Description | Wallaby is an independent Australian API provider for open-weight models — Kimi K3 live now. Prepaid USD credits, per-token metering, itemized receipts. Prompts are never logged or used for training. |
+| Provider Route on LiteLLM | `wallaby/` |
+| Link to Provider Doc | [Wallaby API Docs ↗](https://wallabytoken.com/docs) |
+| Base URL | `https://api.wallabytoken.com/v1` |
+| Supported Operations | [`/chat/completions`](/docs/providers/wallaby#usage---litellm-python-sdk) |
+
+<br />
+
+## What is Wallaby?
+
+Wallaby serves frontier open-weight models behind a standard OpenAI-compatible endpoint:
+- **Open weights only** — Kimi K3 (1M context, reasoning) at launch
+- **Privacy by default** — content logging disabled; billing metadata only (token counts, model, timestamp)
+- **Prepaid USD** — flat rates, balance never expires, itemized receipts
+
+## Required Variables
+
+```python showLineNumbers title="Environment Variables"
+os.environ["WALLABY_API_KEY"] = ""  # your Wallaby API key
+```
+
+Get your API key from [wallabytoken.com](https://wallabytoken.com) — register with an email, top up from $20.
+
+## Usage - LiteLLM Python SDK
+
+### Non-streaming
+
+```python showLineNumbers title="Wallaby Non-streaming Completion"
+import os
+import litellm
+from litellm import completion
+
+os.environ["WALLABY_API_KEY"] = ""  # your Wallaby API key
+
+messages = [{"content": "What is the capital of France?", "role": "user"}]
+
+# Wallaby call
+response = completion(
+    model="wallaby/kimi-k3",
+    messages=messages
+)
+
+print(response)
+```
+
+### Streaming
+
+```python showLineNumbers title="Wallaby Streaming Completion"
+import os
+import litellm
+from litellm import completion
+
+os.environ["WALLABY_API_KEY"] = ""  # your Wallaby API key
+
+messages = [{"content": "Write a short poem about AI", "role": "user"}]
+
+# Wallaby call with streaming
+response = completion(
+    model="wallaby/kimi-k3",
+    messages=messages,
+    stream=True
+)
+
+for chunk in response:
+    print(chunk)
+```
+
+### Reading the reasoning stream
+
+Kimi K3 is a reasoning model — thinking arrives in `reasoning_content`, the answer in `content`:
+
+```python showLineNumbers title="Wallaby Reasoning Stream"
+for chunk in response:
+    delta = chunk.choices[0].delta
+    if getattr(delta, "reasoning_content", None):
+        print(delta.reasoning_content, end="")   # thinking (dim or hide)
+    if delta.content:
+        print(delta.content, end="", flush=True) # the answer
+```
+
+## Usage - LiteLLM Proxy Server
+
+### 1. Save key in your environment
+
+```bash
+export WALLABY_API_KEY=""
+```
+
+### 2. Start the proxy
+
+```yaml
+model_list:
+  - model_name: kimi-k3
+    litellm_params:
+      model: wallaby/kimi-k3
+      api_key: os.environ/WALLABY_API_KEY
+```
+
+## Supported OpenAI Parameters
+
+| Parameter | Type | Description |
+|-----------|------|-------------|
+| `messages` | array | **Required**. Array of message objects with 'role' and 'content' |
+| `model` | string | **Required**. `kimi-k3` |
+| `stream` | boolean | Optional. Enable streaming responses |
+| `max_tokens` / `max_completion_tokens` | integer | Optional. Maximum tokens to generate (thinking counts toward output) |
+| `stop` | string/array | Optional. Stop sequences |
+| `tools` / `tool_choice` | array | Optional. Function calling |
+| `response_format` | object | Optional. `json_object` and `json_schema` (strict) |
+| `temperature` | float | Optional. Accepted and normalized — the serving stack fixes sampling at `1` |
+| `top_p` | float | Optional. Accepted and normalized to `0.95` |
+| `frequency_penalty` / `presence_penalty` | float | Optional. Accepted and normalized to `0` |
+
+## Additional Resources
+
+- [Wallaby API Docs](https://wallabytoken.com/docs)
+- [Pricing (machine-readable)](https://wallabytoken.com/pricing.json)
+- [Status page](https://status.wallabytoken.com/)

sidebars.js

@@ -1257,6 +1257,7 @@ const sidebars = {
         "providers/volcano",
         "providers/voyage",
         "providers/wandb_inference",
+        "providers/wallaby",        
         {
           type: "category",
           label: "WatsonX",