Support DeepSeek-V3.1 tool call by Xu-Wenqing · Pull Request #9446 · sgl-project/sglang

Xu-Wenqing · 2025-08-21T10:05:30Z

Motivation

Support tool call for DeepSeek-V3.1

The tool call format of DeepSeek-V3.1 is different from DeepSeek-V3/R1:
DeepSeek-V3.1: <｜tool▁calls▁begin｜><｜tool▁call▁begin｜>tool_call_name<｜tool▁sep｜>tool_call_arguments<｜tool▁call▁end｜><｜tool▁calls▁end｜>
DeepSeek-R1/V3: <｜tool▁calls▁begin｜><｜tool▁call▁begin｜>function<｜tool▁sep｜>FUNCTION_NAME\n'json\n{"param1": "value1", "param2": "value2"}\n<｜tool▁call▁end｜><｜tool▁calls▁end｜>

So we can't use --tool-call-parser deepseekv3 for this DeepSeek-V3.1, we need a new tool call parser "deepseekv31"

Modifications

Add tool parser "deepseekv31"
Add a new chat template

Accuracy Tests

Test Script (Streaming):

from openai import OpenAI

openai_api_base = ""
openai_api_key = ""

client = OpenAI(
    api_key=openai_api_key,
    base_url=openai_api_base + "/v1",
)

class bcolors:
    HEADER = '\033[95m'
    OKBLUE = '\033[94m'
    OKCYAN = '\033[96m'
    OKGREEN = '\033[92m'
    WARNING = '\033[93m'
    FAIL = '\033[91m'
    ENDC = '\033[0m'
    BOLD = '\033[1m'
    UNDERLINE = '\033[4m'


tools = [
    {
        "type": "function",
        "function": {
            "name": "get_current_temperature",
            "description": "Get current temperature at a location.",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {
                        "type": "string",
                        "description": 'The location to get the temperature for, in the format "City, State, Country".',
                    },
                    "unit": {
                        "type": "string",
                        "enum": ["celsius", "fahrenheit"],
                        "description": 'The unit to return the temperature in. Defaults to "celsius".',
                    },
                },
                "required": ["location"],
            },
            "strict": True
        },
    },
    {
        "type": "function",
        "function": {
            "name": "get_temperature_date",
            "description": "Get temperature at a location and date.",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {
                        "type": "string",
                        "description": 'The location to get the temperature for, in the format "City, State, Country".',
                    },
                    "date": {
                        "type": "string",
                        "description": 'The date to get the temperature for, in the format "Year-Month-Day".',
                    },
                    "unit": {
                        "type": "string",
                        "enum": ["celsius", "fahrenheit"],
                        "description": 'The unit to return the temperature in. Defaults to "celsius".',
                    },
                },
                "required": ["location", "date"],
            },
        },
    },
]


tool_calls_stream = client.chat.completions.create(
    model=client.models.list().data[0].id,
    messages=[
        {
            "role": "system",
            "content": "现在的日期是: 2024-09-30",
        },
        {
            "role": "user",
            "content": "北京今天的天气如何？明天呢？",
        },
    ],
    tools=tools,
    tool_choice="auto",
    stream=True,
    # extra_body={"chat_template_kwargs": {"thinking": True}},
    max_completion_tokens=8192
)

print("reasoning content(Blue) and content(Green):")
chunks = []
for chunk in tool_calls_stream:
    chunks.append(chunk)
    if hasattr(chunk.choices[0].delta, "reasoning_content"):
        reasoning_content = chunk.choices[0].delta.reasoning_content
        if reasoning_content:
            print(bcolors.OKBLUE + reasoning_content, end="", flush=True)
    elif hasattr(chunk.choices[0].delta, "content"):
        content = chunk.choices[0].delta.content
        if content:
            print(bcolors.OKGREEN + content, end="", flush=True)

print(bcolors.ENDC + "\n### end of reasoning content and content. ###\n")

arguments = []
tool_call_idx = -1
for chunk in chunks:
    if chunk.choices[0].delta.tool_calls:
        tool_call = chunk.choices[0].delta.tool_calls[0]

        if tool_call.index != tool_call_idx:
            if tool_call_idx >= 0:
                print(f"streamed tool call arguments: {arguments[tool_call_idx]}")
            tool_call_idx = chunk.choices[0].delta.tool_calls[0].index
            arguments.append("")
        if tool_call.id:
            print(f"streamed tool call id: {tool_call.id} ")

        if tool_call.function:
            if tool_call.function.name:
                print(f"streamed tool call name: {tool_call.function.name}")

            if tool_call.function.arguments:
                arguments[tool_call_idx] += tool_call.function.arguments

if len(arguments):
    print(f"streamed tool call arguments: {arguments[-1]}")

Test Result (Streaming):

reasoning content(Blue) and content(Green):

### end of reasoning content and content. ###

streamed tool call id: call_27ae2109a9134855885ce1e9 
streamed tool call name: get_current_temperature
streamed tool call arguments: {"location": "Beijing, China", "unit": "celsius"}

Test Script (Non-Streaming):

from openai import OpenAI

openai_api_base = ""
openai_api_key = ""

client = OpenAI(
    api_key=openai_api_key,
    base_url=openai_api_base + "/v1",
)


tools = [
    {
        "type": "function",
        "function": {
            "strict": True,
            "name": "get_current_temperature",
            "description": "Get current temperature at a location.",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {
                        "type": "string",
                        "description": 'The location to get the temperature for, in the format "City, State, Country".',
                    },
                    "unit": {
                        "type": "string",
                        "enum": ["celsius", "fahrenheit"],
                        "description": 'The unit to return the temperature in. Defaults to "celsius".',
                    },
                },
                "required": ["location"],
            },
        },
    },
    {
        "type": "function",
        "function": {
            "name": "get_temperature_date",
            "description": "Get temperature at a location and date.",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {
                        "type": "string",
                        "description": 'The location to get the temperature for, in the format "City, State, Country".',
                    },
                    "date": {
                        "type": "string",
                        "description": 'The date to get the temperature for, in the format "Year-Month-Day".',
                    },
                    "unit": {
                        "type": "string",
                        "enum": ["celsius", "fahrenheit"],
                        "description": 'The unit to return the temperature in. Defaults to "celsius".',
                    },
                },
                "required": ["location", "date"],
            },
        },
    },
]


response = client.chat.completions.create(
    model=client.models.list().data[0].id,
    messages=[
        {
            "role": "system",
            "content": "现在的日期是: 2024-09-30",
        },
        {
            "role": "user",
            "content": "北京今天的天气如何？明天呢？",
        },
    ],
    tools=tools,
    tool_choice="auto",
    stream=False,
)

print(response)
tool_calls = response.choices[0].message.tool_calls
for c in tool_calls:
    print(c.function.name, c.function.arguments)

Test Result (Non-Streaming):

ChatCompletion(id='44cbe02d71604830acf43b631e52ab29', choices=[Choice(finish_reason='tool_calls', index=0, logprobs=None, message=ChatCompletionMessage(content='我目前只能帮您查询温度信息，无法提供完整的天气情况（比如降水、风力、湿度等）。\n\n让我为您查询北京当前和明天的温度：', refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=[ChatCompletionMessageToolCall(id='call_2b46341f9f0d48e5b64351c5', function=Function(arguments='{"location": "Beijing, China", "unit": "celsius"}', name='get_current_temperature'), type='function', index=None), ChatCompletionMessageToolCall(id='call_dc0769f36f05479bb9f59107', function=Function(arguments='{"location": "Beijing, China", "date": "2024-01-22", "unit": "celsius"}', name='get_temperature_date'), type='function', index=None)], reasoning_content=None), matched_stop=None)], created=1755842914, model='DeepSeek-V3.1', object='chat.completion', service_tier=None, system_fingerprint=None, usage=CompletionUsage(completion_tokens=90, prompt_tokens=378, total_tokens=468, completion_tokens_details=None, prompt_tokens_details=None, reasoning_tokens=0), metadata={'weight_version': 'default'})
get_current_temperature {"location": "Beijing, China", "unit": "celsius"}
get_temperature_date {"location": "Beijing, China", "date": "2024-01-22", "unit": "celsius"}

Benchmarking and Profiling

Checklist

Format your code according to the Format code with pre-commit.
Add unit tests according to the Run and add unit tests.
Update documentation according to Write documentations.
Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.

Signed-off-by: 许文卿 <xwq391974@alibaba-inc.com>

jimmy-evo · 2025-08-21T13:34:35Z

does it work?

freedomkk-qfeng · 2025-08-21T17:54:25Z

does it work?

not work, here's an example:

request

{
  "messages": [
    {
      "role": "user",
      "content": "Shanghai weather"
    }
  ],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_current_weather",
        "description": "Useful when you want to query the weather of a specified city.",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {
              "type": "string",
              "description": "The city or district, e.g., Beijing, Hangzhou, Yuhang District."
            }
          }
        },
        "required": [
          "location"
        ]
      }
    }
  ],
  "stream": false,
  "model": "deepseek-r1"
}

response

{
  "id": "708ff5439c474fda8f73f2656005a18c",
  "object": "chat.completion",
  "created": 1755798766,
  "model": "deepseek-r1",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "I'll check the weather in Shanghai for you.<｜tool▁calls▁begin｜><｜tool▁call▁begin｜>get_current_weather<｜tool▁sep｜>{\"location\": \"Shanghai\"}<｜tool▁call▁end｜><｜tool▁calls▁end｜>",
        "reasoning_content": null,
        "tool_calls": []
      },
      "logprobs": null,
      "finish_reason": "tool_calls",
      "matched_stop": null
    }
  ],
  "usage": {
    "prompt_tokens": 176,
    "total_tokens": 203,
    "completion_tokens": 27,
    "prompt_tokens_details": null,
    "reasoning_tokens": 0
  },
  "metadata": {
    "weight_version": "default"
  }
}

Xu-Wenqing · 2025-08-22T03:48:49Z

@jinmingyi1998 @freedomkk-qfeng
The tool call format of DeepSeek-V3.1 is different from DeepSeek-V3/R1:
DeepSeek-V3.1: <｜tool▁calls▁begin｜><｜tool▁call▁begin｜>tool_call_name<｜tool▁sep｜>tool_call_arguments<｜tool▁call▁end｜><｜tool▁calls▁end｜>
DeepSeek-R1/V3: <｜tool▁calls▁begin｜><｜tool▁call▁begin｜>function<｜tool▁sep｜>FUNCTION_NAME\n'```json\n{"param1": "value1", "param2": "value2"}\n```<｜tool▁call▁end｜><｜tool▁calls▁end｜>
So we can't use --tool-call-parser deepseekv3 for this DeepSeek-V3.1.

I will write a new tool parser to support DeepSeek-V3.1 CC @CatherineSue @zhyncs @JustinTong0323

Signed-off-by: 许文卿 <xwq391974@alibaba-inc.com>

Xu-Wenqing · 2025-08-22T06:10:13Z

@CatherineSue @JustinTong0323 @zhyncs This PR is ready to review. Added tests results in PR descriptions.

freedomkk-qfeng · 2025-08-25T03:56:00Z

@jinmingyi1998 @freedomkk-qfeng The tool call format of DeepSeek-V3.1 is different from DeepSeek-V3/R1: DeepSeek-V3.1: <｜tool▁calls▁begin｜><｜tool▁call▁begin｜>tool_call_name<｜tool▁sep｜>tool_call_arguments<｜tool▁call▁end｜><｜tool▁calls▁end｜> DeepSeek-R1/V3: <｜tool▁calls▁begin｜><｜tool▁call▁begin｜>function<｜tool▁sep｜>FUNCTION_NAME\n'```json\n{"param1": "value1", "param2": "value2"}\n```<｜tool▁call▁end｜><｜tool▁calls▁end｜> So we can't use --tool-call-parser deepseekv3 for this DeepSeek-V3.1.

I will write a new tool parser to support DeepSeek-V3.1 CC @CatherineSue @zhyncs @JustinTong0323

When thinking mode is disabled, the new parser works!
But when thinking mode is enabled, the parser currently doesn't work correctly

        "reasoning_content": "Hmm, the user is asking about the weather in Shanghai. This is a straightforward request that I can handle with the weather tool. \n\nI need to call the get_current_weather function with Shanghai as the location parameter. The function should return the current weather information for Shanghai.\n\nThe response should be simple and direct - just provide the weather data without unnecessary commentary since the user didn't ask for anything beyond the basic weather information.<｜tool▁calls▁begin｜><｜tool▁call▁begin｜>get_current_weather<｜tool▁sep｜>{\"location\": \"Shanghai\"}<｜tool▁call▁end｜><｜tool▁calls▁end｜>"

Xu-Wenqing · 2025-08-25T04:23:51Z

@jinmingyi1998 @freedomkk-qfeng The tool call format of DeepSeek-V3.1 is different from DeepSeek-V3/R1: DeepSeek-V3.1: <｜tool▁calls▁begin｜><｜tool▁call▁begin｜>tool_call_name<｜tool▁sep｜>tool_call_arguments<｜tool▁call▁end｜><｜tool▁calls▁end｜> DeepSeek-R1/V3: <｜tool▁calls▁begin｜><｜tool▁call▁begin｜>function<｜tool▁sep｜>FUNCTION_NAME\n'```json\n{"param1": "value1", "param2": "value2"}\n```<｜tool▁call▁end｜><｜tool▁calls▁end｜> So we can't use --tool-call-parser deepseekv3 for this DeepSeek-V3.1.
I will write a new tool parser to support DeepSeek-V3.1 CC @CatherineSue @zhyncs @JustinTong0323

When thinking mode is disabled, the new parser works! But when thinking mode is enabled, the parser currently doesn't work correctly
        "reasoning_content": "Hmm, the user is asking about the weather in Shanghai. This is a straightforward request that I can handle with the weather tool. \n\nI need to call the get_current_weather function with Shanghai as the location parameter. The function should return the current weather information for Shanghai.\n\nThe response should be simple and direct - just provide the weather data without unnecessary commentary since the user didn't ask for anything beyond the basic weather information.<｜tool▁calls▁begin｜><｜tool▁call▁begin｜>get_current_weather<｜tool▁sep｜>{\"location\": \"Shanghai\"}<｜tool▁call▁end｜><｜tool▁calls▁end｜>"

@freedomkk-qfeng
https://www.modelscope.cn/models/deepseek-ai/DeepSeek-V3.1 the official doc said tool call only support in non-thinking mode:

examples/chat_template/tool_chat_template_deepseekv31.jinja

wwj-2017-1117 · 2025-08-27T08:38:37Z

--reasoning-parser deepseek-r1
--tool-call-parser deepseekv31 \

curl http://127.0.0.1:30000/v1/chat/completions
-H "Content-Type: application/json"
-d '{
"model": "deepseekv31",
"messages": [
{"role": "user", "content": "What is the weather in San Francisco? Use the weather tool and return a valid JSON tool call."}
],
"tools": [
{
"type": "function",
"function": {
"name": "get_weather",
"description": "Get weather by city",
"parameters": {
"type": "object",
"properties": {"city":{"type":"string"}},
"required": ["city"]
}
}
}
],
"tool_choice": "auto",
"temperature": 0
}'

"tool_calls" still is null

Kevin-XiongC · 2025-08-27T09:15:48Z

--reasoning-parser deepseek-r1

Maybe use --reasoning-parser deepseek-v3?

guodongrui-gdr · 2025-08-27T19:05:36Z

多轮对话调用工具时报错:
openai.InternalServerError: Error code: 500 - {'object': 'error', 'message': 'Internal server error: can only concatenate str (not "dict") to str', 'type': 'InternalServerError', 'param': None, 'code': 500}
测试代码如下:

import json

from openai import OpenAI

client = OpenAI(
    api_key="",
    base_url="http://127.0.0.1:30000/v1" 
)
messages = [
    {
        "role": "user",
        "content": "北京今天的天气如何？",
    },
]
tools = [
    {
        "type": "function",
        "function": {
            "name": "get_current_temperature",
            "description": "Get current temperature at a location.",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {
                        "type": "string",
                        "description": 'The location to get the temperature for, in the format "City, State, Country".',
                    },
                    "unit": {
                        "type": "string",
                        "enum": ["celsius", "fahrenheit"],
                        "description": 'The unit to return the temperature in. Defaults to "celsius".',
                    },
                },
                "required": ["location"],
            },
        },
    },
    {
        "type": "function",
        "function": {
            "name": "get_temperature_date",
            "description": "Get temperature at a location and date.",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {
                        "type": "string",
                        "description": 'The location to get the temperature for, in the format "City, State, Country".',
                    },
                    "date": {
                        "type": "string",
                        "description": 'The date to get the temperature for, in the format "Year-Month-Day".',
                    },
                    "unit": {
                        "type": "string",
                        "enum": ["celsius", "fahrenheit"],
                        "description": 'The unit to return the temperature in. Defaults to "celsius".',
                    },
                },
                "required": ["location", "date"],
            },
        },
    },
]
response = client.chat.completions.create(
    model=client.models.list().data[0].id,
    messages=messages,
    tools=tools,
    stream=False,
    extra_body={"chat_template_kwargs": {"thinking": False}},
)

message = response.choices[0].message
messages.append(message)
if message.tool_calls:
    messages.append({
        'role': 'tool',
        'tool_call_id': message.tool_calls[0].id,
        'content': json.dumps({'response': "25°C"}),
    })

response = client.chat.completions.create(
    model=client.models.list().data[0].id,
    messages=messages,
    tools=tools,
    stream=False,
    extra_body={"chat_template_kwargs": {"thinking": False}},
)
print(response.choices[0].message)

taohui · 2025-08-28T08:42:54Z

多轮对话调用工具时报错: openai.InternalServerError: Error code: 500 - {'object': 'error', 'message': 'Internal server error: can only concatenate str (not "dict") to str', 'type': 'InternalServerError', 'param': None, 'code': 500} 测试代码如下:

import json

from openai import OpenAI

client = OpenAI(
    api_key="",
    base_url="http://127.0.0.1:30000/v1" 
)
messages = [
    {
        "role": "user",
        "content": "北京今天的天气如何？",
    },
]
tools = [
    {
        "type": "function",
        "function": {
            "name": "get_current_temperature",
            "description": "Get current temperature at a location.",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {
                        "type": "string",
                        "description": 'The location to get the temperature for, in the format "City, State, Country".',
                    },
                    "unit": {
                        "type": "string",
                        "enum": ["celsius", "fahrenheit"],
                        "description": 'The unit to return the temperature in. Defaults to "celsius".',
                    },
                },
                "required": ["location"],
            },
        },
    },
    {
        "type": "function",
        "function": {
            "name": "get_temperature_date",
            "description": "Get temperature at a location and date.",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {
                        "type": "string",
                        "description": 'The location to get the temperature for, in the format "City, State, Country".',
                    },
                    "date": {
                        "type": "string",
                        "description": 'The date to get the temperature for, in the format "Year-Month-Day".',
                    },
                    "unit": {
                        "type": "string",
                        "enum": ["celsius", "fahrenheit"],
                        "description": 'The unit to return the temperature in. Defaults to "celsius".',
                    },
                },
                "required": ["location", "date"],
            },
        },
    },
]
response = client.chat.completions.create(
    model=client.models.list().data[0].id,
    messages=messages,
    tools=tools,
    stream=False,
    extra_body={"chat_template_kwargs": {"thinking": False}},
)

message = response.choices[0].message
messages.append(message)
if message.tool_calls:
    messages.append({
        'role': 'tool',
        'tool_call_id': message.tool_calls[0].id,
        'content': json.dumps({'response': "25°C"}),
    })

response = client.chat.completions.create(
    model=client.models.list().data[0].id,
    messages=messages,
    tools=tools,
    stream=False,
    extra_body={"chat_template_kwargs": {"thinking": False}},
)
print(response.choices[0].message)

I also encountered the same problem.

dhh123 · 2025-08-31T17:11:58Z

ExecStart=/apps/miniforge3/envs/sglangpro/bin/python3.13 -m sglang.launch_server
--model-path /llms/DeepSeek-V3.1/
--tool-call-parser deepseekv31
--tokenizer-mode auto
--trust-remote-code
--mem-fraction-static 0.7
--enable-torch-compile
--chunked-prefill-size 4096
--max-running-requests 16
--host 0.0.0.0
--port 40000
--allow-auto-truncate
--tp-size 8
--reasoning-parser deepseek-v3
--chat-template /llms/DeepSeek-V3.1/assets/chat_template.jinja

run deepseek v3.1 with this command . function call not work with openai sdk. in non-thinking @Xu-Wenqing

and tool_chat_template_deepseekv31.jinja this with jinja2 error

this is the wireshark packet
{"chat_template_kwargs":{"thinking":false},"messages":[{"content":" ...........................","role":"user"}],"model":"deepseek-v3.1","tool_choice":"auto","tools":[{"function":{"description":"Get weather of a location, the user should supply a location first.","name":"get_weather","parameters":{"properties":{"location":{"description":"The city and state, e.g. San Francisco, CA","type":"string"}},"required":["location"],"type":"object"}},"type":"function"}]}

T 172.16.8.50:40000 -> 172.16.8.199:50494 [AP] #6
HTTP/1.1 200 OK.
date: Sun, 31 Aug 2025 17:18:01 GMT.
server: uvicorn.
content-length: 1780.
content-type: application/json.
.
{"id":"5f41f79aa53b45cc8a4ea8e78074aecd","object":"chat.completion","created":1756660685,"model":"deepseek-v3.1","choices":[{"index":0,"message":{"role":"assistant","content":"..................................................................AI.......................................................................................................................................................................................\n\n1. ..........................................\n.........................................................\n* ..................App...................................................App........................................................................\n* ..........................................................................................\n * ...............https://weather.cma.cn/\n* ...........................................................................................................................................................................\n\n2. ........................\n....................................................................................................................................................................................................................................\n\n.................................................................................................................................................................................","reasoning_content":null,"tool_calls":null},"logprobs":null,"finish_reason":"stop","matched_stop":1}],"usage":{"prompt_tokens":11,"total_tokens":289,"completion_tokens":278,"prompt_tokens_details":null,"reasoning_tokens":0},"metadata":{"weight_version":"default"}}

pezazc · 2025-09-02T08:01:13Z

Is the multi-round dialogue tool call available now？

JustinTong0323 · 2025-09-02T23:58:40Z

For those who encountered multi-turn FC issue, try this PR:

Update tool_chat_template_deepseekv31.jinja #9895

Signed-off-by: 许文卿 <xwq391974@alibaba-inc.com> Co-authored-by: Xinyuan Tong <xinyuantong.cs@gmail.com>

Xu-Wenqing added 2 commits August 21, 2025 18:04

Add DeepSeek-V3.1 tool call chat template

9234425

Signed-off-by: 许文卿 <xwq391974@alibaba-inc.com>

Add DeepSeek-V3.1 tool call chat template

b9fb19f

Signed-off-by: 许文卿 <xwq391974@alibaba-inc.com>

Xu-Wenqing changed the title ~~[WIP] Add DeepSeek-V3.1 tool call chat template~~ [WIP] Support DeepSeek-V3.1 tool call Aug 22, 2025

Xu-Wenqing added 2 commits August 22, 2025 12:21

Add DeepSeek-V3.1 tool call chat template

7242d0f

Signed-off-by: 许文卿 <xwq391974@alibaba-inc.com>

Merge branch 'main' into add_deepseek_v31_chat_template

443ee7f

Xu-Wenqing changed the title ~~[WIP] Support DeepSeek-V3.1 tool call~~ Support DeepSeek-V3.1 tool call Aug 22, 2025

Xu-Wenqing marked this pull request as ready for review August 22, 2025 06:06

Xu-Wenqing requested a review from CatherineSue as a code owner August 22, 2025 06:06

Xu-Wenqing added 4 commits August 23, 2025 17:53

Merge branch 'main' into add_deepseek_v31_chat_template

debbf69

Merge branch 'main' into add_deepseek_v31_chat_template

8b8b503

Merge branch 'main' into add_deepseek_v31_chat_template

55661f6

Merge branch 'main' into add_deepseek_v31_chat_template

6a46646

Xu-Wenqing mentioned this pull request Aug 25, 2025

[Bug] Function call not work with DeepSeek v3.1 #9552

Closed

5 tasks

Xu-Wenqing added 2 commits August 25, 2025 12:24

Merge branch 'main' into add_deepseek_v31_chat_template

5b4258b

Merge branch 'main' into add_deepseek_v31_chat_template

8d606e7

zhyncs assigned CatherineSue and JustinTong0323 Aug 25, 2025

Xu-Wenqing added 3 commits August 25, 2025 15:37

Merge branch 'main' into add_deepseek_v31_chat_template

c82983a

Merge branch 'main' into add_deepseek_v31_chat_template

6f0a2e7

Merge branch 'main' into add_deepseek_v31_chat_template

7eda5f2

CatherineSue reviewed Aug 26, 2025

View reviewed changes

examples/chat_template/tool_chat_template_deepseekv31.jinja Show resolved Hide resolved

JustinTong0323 added the high priority label Aug 26, 2025

fix lint

a8bf4e3

zhyncs merged commit b9683be into sgl-project:main Aug 27, 2025
80 of 96 checks passed

MahmoudAshraf97 pushed a commit to MahmoudAshraf97/sglang that referenced this pull request Sep 8, 2025

Support DeepSeek-V3.1 tool call (sgl-project#9446)

3bee5ae

Signed-off-by: 许文卿 <xwq391974@alibaba-inc.com> Co-authored-by: Xinyuan Tong <xinyuantong.cs@gmail.com>

Conversation

Xu-Wenqing commented Aug 21, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Motivation

Modifications

Accuracy Tests

Benchmarking and Profiling

Checklist

Uh oh!

jimmy-evo commented Aug 21, 2025

Uh oh!

freedomkk-qfeng commented Aug 21, 2025

Uh oh!

Xu-Wenqing commented Aug 22, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Xu-Wenqing commented Aug 22, 2025

Uh oh!

freedomkk-qfeng commented Aug 25, 2025

Uh oh!

Xu-Wenqing commented Aug 25, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

wwj-2017-1117 commented Aug 27, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Kevin-XiongC commented Aug 27, 2025

Uh oh!

guodongrui-gdr commented Aug 27, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

taohui commented Aug 28, 2025

Uh oh!

dhh123 commented Aug 31, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pezazc commented Sep 2, 2025

Uh oh!

JustinTong0323 commented Sep 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

12 participants

Xu-Wenqing commented Aug 21, 2025 •

edited

Loading

Xu-Wenqing commented Aug 22, 2025 •

edited

Loading

Xu-Wenqing commented Aug 25, 2025 •

edited

Loading

wwj-2017-1117 commented Aug 27, 2025 •

edited

Loading

guodongrui-gdr commented Aug 27, 2025 •

edited

Loading

dhh123 commented Aug 31, 2025 •

edited

Loading

JustinTong0323 commented Sep 2, 2025 •

edited

Loading