--- /tmp/evalscope-sync/src-v191/evalscope/api/model/generate_config.py	2026-07-21 03:18:18.000000000 +0000
+++ /data1/syy/evalscope/evalstone/evalscope/evalscope/api/model/generate_config.py	2026-08-21 03:03:19.541325105 +0000
@@ -64,6 +64,9 @@
     max_tokens: Optional[int] = Field(default=None)
     """The maximum number of tokens that can be generated in the completion (default is model specific)."""
 
+    max_completion_tokens: Optional[int] = Field(default=None)
+    """An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens. Used by OpenAI reasoning models and some proxies."""
+
     top_p: Optional[float] = Field(default=None)
     """An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass."""