Use allow auto truncate in the OpenAI API endpoint (#15369)

2025-12-25 18:45:57 -08:00
parent 0271fc3456
commit a91e072f33
1 changed files with 1 additions and 1 deletions
--- a/python/sglang/srt/entrypoints/openai/serving_chat.py
+++ b/python/sglang/srt/entrypoints/openai/serving_chat.py
@@ -133,7 +133,7 @@ class OpenAIServingChat(OpenAIServingBase):
            max_output_tokens
            and server_context_length
            and max_output_tokens > server_context_length
-        ):
+        ) and not self.tokenizer_manager.server_args.allow_auto_truncate:
            return (
                f"max_completion_tokens is too large: {max_output_tokens}."
                f"This model supports at most {server_context_length} completion tokens."