[1/N] API Server (Remove Proxy) (#11529)

vllm/entrypoints/openai/api_server.py

@@ -585,9 +585,15 @@ def build_app(args: Namespace) -> FastAPI:
                                     status_code=401)
             return await call_next(request)
 
+    if args.enable_request_id_headers:
+        logger.warning(
+            "CAUTION: Enabling X-Request-Id headers in the API Server. "
+            "This can harm performance at high QPS.")
+
         @app.middleware("http")
         async def add_request_id(request: Request, call_next):
-        request_id = request.headers.get("X-Request-Id") or uuid.uuid4().hex
+            request_id = request.headers.get(
+                "X-Request-Id") or uuid.uuid4().hex
             response = await call_next(request)
             response.headers["X-Request-Id"] = request_id
             return response

vllm/entrypoints/openai/cli_args.py

@@ -196,7 +196,11 @@ def make_arg_parser(parser: FlexibleArgumentParser) -> FlexibleArgumentParser:
         action="store_true",
         help="If specified, will run the OpenAI frontend server in the same "
         "process as the model serving engine.")
-
+    parser.add_argument(
+        "--enable-request-id-headers",
+        action="store_true",
+        help="If specified, API server will add X-Request-Id header to "
+        "responses. Caution: this hurts performance at high QPS.")
     parser.add_argument(
         "--enable-auto-tool-choice",
         action="store_true",