biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Refactor] Simplify chat_completion_full_generator for tool parsers (#35634)

Browse Source 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Wentao Ye
2026-03-09 11:33:46 -04:00
committed by GitHub
parent be292b7c14
commit 941e52c298
1 changed files with 1 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
vllm/entrypoints/openai/chat_completion/serving.py
View File

@@ -1463,17 +1463,7 @@ class OpenAIServingChat(OpenAIServing):
tool_call_class = (
MistralToolCall if is_mistral_tokenizer(tokenizer) else ToolCall
)
if self.use_harmony:
# Harmony models already have parsed content and tool_calls
# through parse_chat_output. Respect its output directly.
message = ChatMessage(
role=role,
reasoning=reasoning,
content=content,
tool_calls=tool_calls if tool_calls else [],
)
elif (not self.enable_auto_tools or not self.tool_parser) and (
if (not self.enable_auto_tools or not self.tool_parser) and (
not isinstance(request.tool_choice, ChatCompletionNamedToolChoiceParam)
and request.tool_choice != "required"
):