From 170d1f218a4ea33067aabf12bafc4b51b255ca18 Mon Sep 17 00:00:00 2001 From: Chang Su Date: Thu, 1 May 2025 21:28:57 -0700 Subject: [PATCH] feat: Refactor DeepSeekV3 function call (#5908) --- docs/references/deepseek.md | 7 +- .../tool_chat_template_deepseekv3.jinja | 91 +++++++++++++++++++ python/sglang/srt/openai_api/adapter.py | 26 ------ 3 files changed, 95 insertions(+), 29 deletions(-) create mode 100644 examples/chat_template/tool_chat_template_deepseekv3.jinja diff --git a/docs/references/deepseek.md b/docs/references/deepseek.md index 80de28ca8..a7d69e7eb 100644 --- a/docs/references/deepseek.md +++ b/docs/references/deepseek.md @@ -171,10 +171,10 @@ See [Separate Reasoning](https://docs.sglang.ai/backend/separate_reasoning.html) ### Function calling for DeepSeek Models -Add arguments `--tool-call-parser deepseekv3` to enable this feature. For example (running on 1 * H20 node): +Add arguments `--tool-call-parser deepseekv3` and `--chat-template ./examples/chat_template/tool_chat_template_deepseekv3.jinja`(recommended) to enable this feature. For example (running on 1 * H20 node): ``` -python3 -m sglang.launch_server --model deepseek-ai/DeepSeek-V3-0324 --tp 8 --port 30000 --host 0.0.0.0 --mem-fraction-static 0.9 --disable-cuda-graph --tool-call-parser deepseekv3 +python3 -m sglang.launch_server --model deepseek-ai/DeepSeek-V3-0324 --tp 8 --port 30000 --host 0.0.0.0 --mem-fraction-static 0.9 --disable-cuda-graph --tool-call-parser deepseekv3 --chat-template ./examples/chat_template/tool_chat_template_deepseekv3.jinja ``` Sample Request: @@ -188,7 +188,7 @@ curl "http://127.0.0.1:30000/v1/chat/completions" \ Expected Response ``` -{"id": "62af80528930423a82c806651ec66e7c", "object": "chat.completion", "created": 1744431333, "model": "deepseek-ai/DeepSeek-V3-0324", "choices": [{"index": 0, "message": {"role": "assistant", "content": null, "reasoning_content": null, "tool_calls": [{"id": "0", "type": "function", "function": {"name": "query_weather", "arguments": "{\\"city\\": \\"Guangzhou\\"}"}}]}, "logprobs": null, "finish_reason": "tool_calls", "matched_stop": null}], "usage": {"prompt_tokens": 118, "total_tokens": 140, "completion_tokens": 22, "prompt_tokens_details": null}} +{"id":"6501ef8e2d874006bf555bc80cddc7c5","object":"chat.completion","created":1745993638,"model":"deepseek-ai/DeepSeek-V3-0324","choices":[{"index":0,"message":{"role":"assistant","content":null,"reasoning_content":null,"tool_calls":[{"id":"0","index":null,"type":"function","function":{"name":"query_weather","arguments":"{\"city\": \"Qingdao\"}"}}]},"logprobs":null,"finish_reason":"tool_calls","matched_stop":null}],"usage":{"prompt_tokens":116,"total_tokens":138,"completion_tokens":22,"prompt_tokens_details":null}} ``` Sample Streaming Request: @@ -215,6 +215,7 @@ The client needs to concatenate all arguments fragments to reconstruct the compl ``` Important Notes: 1. Use a lower `"temperature"` value for better results. +2. To receive more consistent tool call results, it is recommended to use `--chat-template examples/chat_template/tool_chat_template_deepseekv3.jinja`. It provides an improved unified prompt. diff --git a/examples/chat_template/tool_chat_template_deepseekv3.jinja b/examples/chat_template/tool_chat_template_deepseekv3.jinja new file mode 100644 index 000000000..dde922d30 --- /dev/null +++ b/examples/chat_template/tool_chat_template_deepseekv3.jinja @@ -0,0 +1,91 @@ +{% if not add_generation_prompt is defined %} + {% set add_generation_prompt = false %} +{% endif %} + +{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='', is_first_sp=true, is_last_user=false) %} +{%- for message in messages %} + {%- if message['role'] == 'system' %} + {%- if ns.is_first_sp %} + {% set ns.system_prompt = ns.system_prompt + message['content'] %} + {% set ns.is_first_sp = false %} + {%- else %} + {% set ns.system_prompt = ns.system_prompt + '\n\n' + message['content'] %} + {%- endif %} + {%- endif %} +{%- endfor %} + +{# --- Append tool descriptions if tools are defined --- #} +{% if tools is defined and tools is not none %} + {% set tool_ns = namespace(text='You are a helpful assistant with tool calling capabilities. ' + 'When a tool call is needed, you MUST use the following format to issue the call:\n' + '<｜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｜>\n\n' + 'Make sure the JSON is valid.' + '## Tools\n\n### Function\n\nYou have the following functions available:\n\n') %} + {% for tool in tools %} + {% set tool_ns.text = tool_ns.text + '- `' + tool['name'] + '`:\n```json\n' + (tool | tojson) + '\n```\n' %} + {% endfor %} + {% set ns.system_prompt = ns.system_prompt + '\n\n' + tool_ns.text %} +{% endif %} + +{{ bos_token }} +{{ ns.system_prompt }} + +{%- for message in messages %} + {%- if message['role'] == 'user' %} + {%- set ns.is_tool = false -%} + {%- set ns.is_first = false -%} + {%- set ns.is_last_user = true -%} + {{'<｜User｜>' + message['content'] + '<｜Assistant｜>'}} + {%- endif %} + {%- if message['role'] == 'assistant' and message['tool_calls'] is defined and message['tool_calls'] is not none %} + {%- set ns.is_last_user = false -%} + {%- if ns.is_tool %} + {{'<｜tool▁outputs▁end｜>'}} + {%- endif %} + {%- set ns.is_first = false %} + {%- set ns.is_tool = false -%} + {%- set ns.is_output_first = true %} + {%- for tool in message['tool_calls'] %} + {%- if not ns.is_first %} + {%- if message['content'] is none %} + {{'<｜tool▁calls▁begin｜><｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\n' + '```json' + '\n' + tool['function']['arguments'] + '\n' + '```' + '<｜tool▁call▁end｜>'}} + {%- else %} + {{message['content'] + '<｜tool▁calls▁begin｜><｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\n' + '```json' + '\n' + tool['function']['arguments'] + '\n' + '```' + '<｜tool▁call▁end｜>'}} + {%- endif %} + {%- set ns.is_first = true -%} + {%- else %} + {{'\n' + '<｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\n' + '```json' + '\n' + tool['function']['arguments'] + '\n' + '```' + '<｜tool▁call▁end｜>'}} + {%- endif %} + {%- endfor %} + {{'<｜tool▁calls▁end｜><｜end▁of▁sentence｜>'}} + {%- endif %} + {%- if message['role'] == 'assistant' and (message['tool_calls'] is not defined or message['tool_calls'] is none)%} + {%- set ns.is_last_user = false -%} + {%- if ns.is_tool %} + {{'<｜tool▁outputs▁end｜>' + message['content'] + '<｜end▁of▁sentence｜>'}} + {%- set ns.is_tool = false -%} + {%- else %} + {% set content = message['content'] %} + {{content + '<｜end▁of▁sentence｜>'}} + {%- endif %} + {%- endif %} + {%- if message['role'] == 'tool' %} + {%- set ns.is_last_user = false -%} + {%- set ns.is_tool = true -%} + {%- if ns.is_output_first %} + {{ 'Use the results below to formulate an answer to the user question unless additional information is needed.' }} + {{'<｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}} + {%- set ns.is_output_first = false %} + {%- else %} + {{'\n<｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}} + {%- endif %} + {%- endif %} +{%- endfor -%} + +{% if ns.is_tool %} + {{"<｜tool▁outputs▁end｜>"}} +{% endif %} +{% if add_generation_prompt and not ns.is_last_user and not ns.is_tool %} + {{'<｜Assistant｜>'}} +{% endif %} diff --git a/python/sglang/srt/openai_api/adapter.py b/python/sglang/srt/openai_api/adapter.py index 74333e0ce..d4640e23b 100644 --- a/python/sglang/srt/openai_api/adapter.py +++ b/python/sglang/srt/openai_api/adapter.py @@ -966,32 +966,6 @@ def v1_chat_generate_request( if chat_template_name is None: openai_compatible_messages = [] - if ( - tools - and tokenizer_manager.server_args.tool_call_parser == "deepseekv3" - ): - # add function call prompt to deepseekv3 - openai_compatible_messages.append( - { - "role": "system", - "content": """You are a helpful Assistant. - ## Tools - ### Function - You have the following functions available: - """ - + "".join( - [ - f""" - - `{tool['name']}`: - ```json - {json.dumps(tool)} - ``` - """ - for tool in tools - ] - ), - } - ) for message in request.messages: if message.content is None: