From da2e5d6546755507fc3b893d9c68e62f07c77311 Mon Sep 17 00:00:00 2001 From: Lianmin Zheng Date: Tue, 9 Jul 2024 01:52:55 -0700 Subject: [PATCH] Fix the default argument of OpenAI Chat completion (#605) --- README.md | 1 - python/sglang/srt/openai_protocol.py | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/README.md b/README.md index 5861c7a1c..2ac666c6b 100644 --- a/README.md +++ b/README.md @@ -42,7 +42,6 @@ pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.3/ git clone https://github.com/sgl-project/sglang.git cd sglang -pip install --upgrade pip pip install -e "python[all]" # Install FlashInfer CUDA kernels diff --git a/python/sglang/srt/openai_protocol.py b/python/sglang/srt/openai_protocol.py index ce37097a7..dfe58e857 100644 --- a/python/sglang/srt/openai_protocol.py +++ b/python/sglang/srt/openai_protocol.py @@ -134,7 +134,7 @@ class ChatCompletionRequest(BaseModel): logit_bias: Optional[Dict[str, float]] = None logprobs: Optional[bool] = False top_logprobs: Optional[int] = None - max_tokens: Optional[int] = None + max_tokens: Optional[int] = 16 n: Optional[int] = 1 presence_penalty: Optional[float] = 0.0 response_format: Optional[ResponseFormat] = None