diff --git a/fastdeploy/entrypoints/openai/serving_chat.py b/fastdeploy/entrypoints/openai/serving_chat.py
index a5af82ff5..388a81dd9 100644
--- a/fastdeploy/entrypoints/openai/serving_chat.py
+++ b/fastdeploy/entrypoints/openai/serving_chat.py
@@ -87,7 +87,7 @@ class OpenAIServingChat:
                 f"Only master node can accept completion request, please send request to master node: {self.master_ip}"
             )
             api_server_logger.error(err_msg)
-            return ErrorResponse(error=ErrorInfo(message=err_msg, type=ErrorType.SERVER_ERROR))
+            return ErrorResponse(error=ErrorInfo(message=err_msg, type=ErrorType.INTERNAL_ERROR))
 
         if self.models:
             is_supported, request.model = self.models.is_supported_model(request.model)
@@ -95,7 +95,7 @@ class OpenAIServingChat:
                 err_msg = f"Unsupported model: [{request.model}], support [{', '.join([x.name for x in self.models.model_paths])}] or default"
                 api_server_logger.error(err_msg)
                 return ErrorResponse(
-                    error=ErrorInfo(message=err_msg, type=ErrorType.SERVER_ERROR, code=ErrorCode.MODEL_NOT_SUPPORT)
+                    error=ErrorInfo(message=err_msg, type=ErrorType.INTERNAL_ERROR, code=ErrorCode.MODEL_NOT_SUPPORT)
                 )
 
         try:
@@ -145,7 +145,7 @@ class OpenAIServingChat:
                 except Exception as e:
                     error_msg = f"request[{request_id}]full generator error: {str(e)}, {str(traceback.format_exc())}"
                     api_server_logger.error(error_msg)
-                    return ErrorResponse(error=ErrorInfo(message=error_msg, type=ErrorType.SERVER_ERROR))
+                    return ErrorResponse(error=ErrorInfo(message=error_msg, type=ErrorType.INTERNAL_ERROR))
         except Exception as e:
             error_msg = (
                 f"request[{request_id}] waiting error: {str(e)}, {str(traceback.format_exc())}, "
@@ -158,7 +158,7 @@ class OpenAIServingChat:
 
     def _create_streaming_error_response(self, message: str) -> str:
         api_server_logger.error(message)
-        error_response = ErrorResponse(error=ErrorInfo(message=message, type=ErrorType.SERVER_ERROR))
+        error_response = ErrorResponse(error=ErrorInfo(message=message, type=ErrorType.INTERNAL_ERROR))
         return error_response.model_dump_json()
 
     async def chat_completion_stream_generator(
diff --git a/fastdeploy/entrypoints/openai/serving_completion.py b/fastdeploy/entrypoints/openai/serving_completion.py
index b761dfeac..6bbeba10c 100644
--- a/fastdeploy/entrypoints/openai/serving_completion.py
+++ b/fastdeploy/entrypoints/openai/serving_completion.py
@@ -64,14 +64,14 @@ class OpenAIServingCompletion:
                 f"Only master node can accept completion request, please send request to master node: {self.master_ip}"
             )
             api_server_logger.error(err_msg)
-            return ErrorResponse(error=ErrorInfo(message=err_msg, type=ErrorType.SERVER_ERROR))
+            return ErrorResponse(error=ErrorInfo(message=err_msg, type=ErrorType.INTERNAL_ERROR))
         if self.models:
             is_supported, request.model = self.models.is_supported_model(request.model)
             if not is_supported:
                 err_msg = f"Unsupported model: [{request.model}], support [{', '.join([x.name for x in self.models.model_paths])}] or default"
                 api_server_logger.error(err_msg)
                 return ErrorResponse(
-                    error=ErrorInfo(message=err_msg, type=ErrorType.SERVER_ERROR, code=ErrorCode.MODEL_NOT_SUPPORT)
+                    error=ErrorInfo(message=err_msg, type=ErrorType.INTERNAL_ERROR, code=ErrorCode.MODEL_NOT_SUPPORT)
                 )
         created_time = int(time.time())
         if request.user is not None:
@@ -115,7 +115,7 @@ class OpenAIServingCompletion:
         except Exception as e:
             error_msg = f"OpenAIServingCompletion create_completion: {e}, {str(traceback.format_exc())}"
             api_server_logger.error(error_msg)
-            return ErrorResponse(error=ErrorInfo(message=error_msg, type=ErrorType.SERVER_ERROR))
+            return ErrorResponse(error=ErrorInfo(message=error_msg, type=ErrorType.INTERNAL_ERROR))
 
         if request_prompt_ids is not None:
             request_prompts = request_prompt_ids
@@ -189,12 +189,12 @@ class OpenAIServingCompletion:
                         f"OpenAIServingCompletion completion_full_generator error: {e}, {str(traceback.format_exc())}"
                     )
                     api_server_logger.error(error_msg)
-                    return ErrorResponse(error=ErrorInfo(message=error_msg, type=ErrorType.SERVER_ERROR))
+                    return ErrorResponse(error=ErrorInfo(message=error_msg, type=ErrorType.INTERNAL_ERROR))
 
         except Exception as e:
             error_msg = f"OpenAIServingCompletion create_completion error: {e}, {str(traceback.format_exc())}"
             api_server_logger.error(error_msg)
-            return ErrorResponse(error=ErrorInfo(message=error_msg, type=ErrorType.SERVER_ERROR))
+            return ErrorResponse(error=ErrorInfo(message=error_msg, type=ErrorType.INTERNAL_ERROR))
 
     async def completion_full_generator(
         self,
diff --git a/fastdeploy/entrypoints/openai/serving_models.py b/fastdeploy/entrypoints/openai/serving_models.py
index 74f925947..fc1507e5f 100644
--- a/fastdeploy/entrypoints/openai/serving_models.py
+++ b/fastdeploy/entrypoints/openai/serving_models.py
@@ -87,7 +87,7 @@ class OpenAIServingModels:
                 f"Only master node can accept models request, please send request to master node: {self.master_ip}"
             )
             api_server_logger.error(err_msg)
-            return ErrorResponse(error=ErrorInfo(message=err_msg, type=ErrorType.SERVER_ERROR))
+            return ErrorResponse(error=ErrorInfo(message=err_msg, type=ErrorType.INTERNAL_ERROR))
         model_infos = [
             ModelInfo(
                 id=model.name, max_model_len=self.max_model_len, root=model.model_path, permission=[ModelPermission()]
diff --git a/tests/utils/test_exception_handler.py b/tests/utils/test_exception_handler.py
index c5e2d0855..e9bd034ea 100644
--- a/tests/utils/test_exception_handler.py
+++ b/tests/utils/test_exception_handler.py
@@ -8,6 +8,10 @@ from fastapi.responses import JSONResponse
 from fastdeploy.utils import ErrorCode, ExceptionHandler, ParameterError
 
 
+class DummyRequest:
+    url = "http://testserver/test"
+
+
 class TestParameterError(unittest.TestCase):
     def test_parameter_error_init(self):
         exc = ParameterError("param1", "error message")
@@ -30,7 +34,8 @@ class TestExceptionHandler(unittest.IsolatedAsyncioTestCase):
     async def test_handle_request_validation_missing_messages(self):
         """缺少 messages 参数时，应返回 missing_required_parameter"""
         exc = RequestValidationError([{"loc": ("body", "messages"), "msg": "Field required", "type": "missing"}])
-        resp: JSONResponse = await ExceptionHandler.handle_request_validation_exception(None, exc)
+        dummy_request = DummyRequest()
+        resp: JSONResponse = await ExceptionHandler.handle_request_validation_exception(dummy_request, exc)
         data = json.loads(resp.body.decode())
         self.assertEqual(resp.status_code, HTTPStatus.BAD_REQUEST)
         self.assertEqual(data["error"]["param"], "messages")
@@ -42,7 +47,8 @@ class TestExceptionHandler(unittest.IsolatedAsyncioTestCase):
         exc = RequestValidationError(
             [{"loc": ("body", "top_p"), "msg": "Input should be less than or equal to 1", "type": "value_error"}]
         )
-        resp: JSONResponse = await ExceptionHandler.handle_request_validation_exception(None, exc)
+        dummy_request = DummyRequest()
+        resp: JSONResponse = await ExceptionHandler.handle_request_validation_exception(dummy_request, exc)
         data = json.loads(resp.body.decode())
         self.assertEqual(resp.status_code, HTTPStatus.BAD_REQUEST)
         self.assertEqual(data["error"]["param"], "top_p")