A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 6, 21:45 UTC
[Refactor] Remove deadcode in Responses API serving (#36726)
Signed-off-by: sfeng33 <4florafeng@gmail.com> Co-authored-by: Signed-off-by: yewentao256 <zhyanwentao@126.com>
Flora Feng committed
Mar 11, 2026 at 03:11 UTC
d5080aeaa4d80f285d436ef66159fb2de4ffd3f7
1 file changed
-23
vllm/entrypoints/openai/responses/serving.py
-23
@@ -1102,7 +1102,6 @@ class OpenAIServingResponses(OpenAIServing):
1102
event_deque: deque[StreamingResponsesResponse] = deque()
1103
new_event_signal = asyncio.Event()
1104
self.event_store[request.request_id] = (event_deque, new_event_signal)
1105
- response = None
1105
generator = self.responses_stream_generator(request, *args, **kwargs)
1106
try:
1107
async for event in generator:
@@ -1111,15 +1110,6 @@ class OpenAIServingResponses(OpenAIServing):
1110
finally:
1111
new_event_signal.set()
1112
1114
- if response is not None and isinstance(response, ErrorResponse):
1115
- # If the request has failed, update the status to "failed".
1116
- response_id = request.request_id
1117
- async with self.response_store_lock:
1118
- stored_response = self.response_store.get(response_id)
1119
- assert stored_response is not None
1120
- if stored_response.status not in ("completed", "cancelled"):
1121
- stored_response.status = "failed"
1122
-
1113
async def _run_background_request(
1114
self,
1115
request: ResponsesRequest,
@@ -1226,19 +1216,6 @@ class OpenAIServingResponses(OpenAIServing):
1216
param="response_id",
1217
)
1218
1229
- def _make_store_not_supported_error(self) -> ErrorResponse:
1230
- return self.create_error_response(
1231
- err_type="invalid_request_error",
1232
- message=(
1233
- "`store=True` (default) is not supported. Please set "
1234
- "`store=False` in Responses API or set "
1235
- "`VLLM_ENABLE_RESPONSES_API_STORE=1` in the env var when "
1236
- "starting the vLLM server."
1237
- ),
1238
- status_code=HTTPStatus.BAD_REQUEST,
1239
- param="store",
1240
- )
1241
-
1219
async def _process_simple_streaming_events(
1220
self,
1221
request: ResponsesRequest,