biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Model][6/N] Improve all pooling task | Support chunked prefill with ALL pooling (#27145)

Browse Source 
Signed-off-by: wang.yuqi <noooop@126.com>
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
This commit is contained in:
wang.yuqi
2025-12-04 21:44:15 +08:00
committed by GitHub
parent 1b7c7f5159
commit 74c4d80c6c
15 changed files with 224 additions and 93 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
tests/entrypoints/pooling/classify/test_online.py
View File

@@ -255,21 +255,21 @@ async def test_pooling_classify(server: RemoteOpenAIServer, model_name: str):
@pytest.mark.asyncio
@pytest.mark.parametrize("model_name", [MODEL_NAME])
async def test_pooling_token_classify(server: RemoteOpenAIServer, model_name: str):
# token_classify uses ALL pooling, which does not support chunked prefill.
task = "token_classify"
input_text = ["This product was excellent and exceeded my expectations"]
response = requests.post(
server.url_for("pooling"),
json={
"model": model_name,
"input": "test",
"input": input_text,
"encoding_format": "float",
"task": task,
},
)
assert response.json()["error"]["type"] == "BadRequestError"
assert response.json()["error"]["message"].startswith(
f"Task {task} is not supported"
)
poolings = PoolingResponse.model_validate(response.json())
assert len(poolings.data) == 1
assert len(poolings.data[0].data) == 8
assert len(poolings.data[0].data[0]) == 2
@pytest.mark.asyncio