diff --git a/python/sglang/srt/openai_api/adapter.py b/python/sglang/srt/openai_api/adapter.py
index 58b0d5fd9..911969859 100644
--- a/python/sglang/srt/openai_api/adapter.py
+++ b/python/sglang/srt/openai_api/adapter.py
@@ -1099,7 +1099,7 @@ def v1_chat_generate_request(
 
         sampling_params = {
             "temperature": request.temperature,
-            "max_new_tokens": request.max_tokens,
+            "max_new_tokens": request.max_tokens or request.max_completion_tokens,
             "min_new_tokens": request.min_tokens,
             "stop": stop,
             "stop_token_ids": request.stop_token_ids,
diff --git a/python/sglang/srt/openai_api/protocol.py b/python/sglang/srt/openai_api/protocol.py
index 9075a1a69..39e25a57c 100644
--- a/python/sglang/srt/openai_api/protocol.py
+++ b/python/sglang/srt/openai_api/protocol.py
@@ -320,7 +320,16 @@ class ChatCompletionRequest(BaseModel):
     logit_bias: Optional[Dict[str, float]] = None
     logprobs: bool = False
     top_logprobs: Optional[int] = None
-    max_tokens: Optional[int] = None
+    max_tokens: Optional[int] = Field(
+        default=None,
+        deprecated="max_tokens is deprecated in favor of the max_completion_tokens field",
+        description="The maximum number of tokens that can be generated in the chat completion. ",
+    )
+    max_completion_tokens: Optional[int] = Field(
+        default=None,
+        description="The maximum number of completion tokens for a chat completion request, "
+        "including visible output tokens and reasoning tokens. Input tokens are not included. ",
+    )
     n: int = 1
     presence_penalty: float = 0.0
     response_format: Optional[Union[ResponseFormat, StructuralTagResponseFormat]] = None