[Pooling] Honor request_id from request bodies (#55665)
Signed-off-by: Taneem Ibrahim <taneem.ibrahim@gmail.com>
Taneem Ibrahim committed
Sep 7, 2026 at 18:55 UTC
fdfa0a1659a4a75de607219f9a7004d09928d572
2 files changed
+7
-1
tests/entrypoints/pooling/token_embed/test_online.py
+3
@@ -54,6 +54,7 @@ def server():
54
@pytest.mark.parametrize("model_name", [MODEL_NAME])
55
async def test_pooling_token_embed(server: RemoteOpenAIServer, model_name: str):
56
task = "token_embed"
57
+ request_id = "client-request-id"
58
response = requests.post(
59
server.url_for("pooling"),
60
json={
@@ -61,11 +62,13 @@ async def test_pooling_token_embed(server: RemoteOpenAIServer, model_name: str):
62
"input": input_text,
63
"encoding_format": "float",
64
"task": task,
65
+ "request_id": request_id,
66
},
67
)
68
69
poolings = PoolingResponse.model_validate(response.json())
70
71
+ assert poolings.id == f"pooling-{request_id}"
72
assert len(poolings.data) == 1
73
assert len(poolings.data[0].data) == len(input_tokens)
74
assert len(poolings.data[0].data[0]) == 384
vllm/entrypoints/pooling/base/serving.py
+4
-1
@@ -109,7 +109,10 @@ class PoolingBaseServing(ABC, BaseServing):
109
raw_request: Request | None = None,
110
):
111
model_name = self.models.model_name()
112
- request_id = f"{self.request_id_prefix}-{self._base_request_id(raw_request)}"
112
+ base_request_id = self._base_request_id(
113
+ raw_request, getattr(request, "request_id", None)
114
+ )
115
+ request_id = f"{self.request_id_prefix}-{base_request_id}"
116
await self._check_model(request)
117
118
pooling_params = io_processor.create_pooling_params(request)