@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / c77eb8a33c
Fork 0

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] Set temperature=0.7 in test_guided_choice_chat (#11264)

Michael Goin committed Dec 17, 2024 at 19:34 UTC c77eb8a33ceb62858d951ffef87ae626a0d09973
1 file changed +2
tests/entrypoints/openai/test_chat.py
+2
@@ -482,6 +482,7 @@ async def test_guided_choice_chat(client: openai.AsyncOpenAI,
482 model=MODEL_NAME,
483 messages=messages,
484 max_completion_tokens=10,
485 + temperature=0.7,
486 extra_body=dict(guided_choice=sample_guided_choice,
487 guided_decoding_backend=guided_decoding_backend))
488 choice1 = chat_completion.choices[0].message.content
@@ -496,6 +497,7 @@ async def test_guided_choice_chat(client: openai.AsyncOpenAI,
497 model=MODEL_NAME,
498 messages=messages,
499 max_completion_tokens=10,
500 + temperature=0.7,
501 extra_body=dict(guided_choice=sample_guided_choice,
502 guided_decoding_backend=guided_decoding_backend))
503 choice2 = chat_completion.choices[0].message.content