A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 16, 04:30 UTC
[Bugfix] Fix the default value for temperature in ChatCompletionRequest (#11219)
yansh97 committed
Dec 16, 2024 at 16:15 UTC
17138af7c45eba3aba3e9b84a3852b4ba81e460f
1 file changed
+1
-1
vllm/entrypoints/openai/protocol.py
+1
-1
@@ -211,7 +211,7 @@ class ChatCompletionRequest(OpenAIBaseModel):
211
stop: Optional[Union[str, List[str]]] = Field(default_factory=list)
212
stream: Optional[bool] = False
213
stream_options: Optional[StreamOptions] = None
214
- temperature: Optional[float] = 0.7
214
+ temperature: Optional[float] = 1.0
215
top_p: Optional[float] = 1.0
216
tools: Optional[List[ChatCompletionToolsParam]] = None
217
tool_choice: Optional[Union[Literal["none"], Literal["auto"],