diff --git a/fastdeploy/entrypoints/openai/serving_chat.py b/fastdeploy/entrypoints/openai/serving_chat.py
index df074771c..a71c6d14f 100644
--- a/fastdeploy/entrypoints/openai/serving_chat.py
+++ b/fastdeploy/entrypoints/openai/serving_chat.py
@@ -182,7 +182,7 @@ class OpenAIServingChat:
             if request.max_streaming_response_tokens is not None
             else (request.metadata or {}).get("max_streaming_response_tokens", 1)
         )  # dierctly passed & passed in metadata
-
+        max_streaming_response_tokens = max(max_streaming_response_tokens, 1)
         enable_thinking = request.chat_template_kwargs.get("enable_thinking") if request.chat_template_kwargs else None
         if enable_thinking is None:
             enable_thinking = request.metadata.get("enable_thinking") if request.metadata else None
@@ -370,11 +370,6 @@ class OpenAIServingChat:
                             api_server_logger.info(f"Chat Streaming response last send: {chunk.model_dump_json()}")
                         choices = []
 
-                if choices:
-                    chunk.choices = choices
-                    yield f"data: {chunk.model_dump_json(exclude_unset=True)}\n\n"
-                    choices = []
-
             if include_usage:
                 completion_tokens = previous_num_tokens
                 usage = UsageInfo(
diff --git a/fastdeploy/entrypoints/openai/serving_completion.py b/fastdeploy/entrypoints/openai/serving_completion.py
index 059c22da3..c42c81839 100644
--- a/fastdeploy/entrypoints/openai/serving_completion.py
+++ b/fastdeploy/entrypoints/openai/serving_completion.py
@@ -331,6 +331,7 @@ class OpenAIServingCompletion:
                 if request.max_streaming_response_tokens is not None
                 else (request.suffix or {}).get("max_streaming_response_tokens", 1)
             )  # dierctly passed & passed in suffix
+            max_streaming_response_tokens = max(max_streaming_response_tokens, 1)
             choices = []
             chunk = CompletionStreamResponse(
                 id=request_id,
@@ -461,10 +462,6 @@ class OpenAIServingCompletion:
                             )
                             yield f"data: {usage_chunk.model_dump_json(exclude_unset=True)}\n\n"
                         api_server_logger.info(f"Completion Streaming response last send: {chunk.model_dump_json()}")
-                if choices:
-                    chunk.choices = choices
-                    yield f"data: {chunk.model_dump_json(exclude_unset=True)}\n\n"
-                    choices = []
 
         except Exception as e:
             api_server_logger.error(f"Error in completion_stream_generator: {e}, {str(traceback.format_exc())}")