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

[Bugfix][CI/Build] Fix failing pooling models test due to Triton kernel accuracy diff (#31776)

Browse Source 
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
This commit is contained in:
Isotr0py
2026-01-06 16:44:22 +08:00
committed by GitHub
parent 7101e0851f
commit ee2e69d6cd
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tests/models/language/pooling/test_token_classification.py
View File

@@ -78,7 +78,7 @@ def test_modernbert_models(
for hf_output, vllm_output in zip(hf_outputs, vllm_outputs):
hf_output = hf_output.detach().clone().cpu().float()
vllm_output = vllm_output.detach().clone().cpu().float()
assert torch.allclose(hf_output, vllm_output, atol=1e-2)
torch.testing.assert_close(hf_output, vllm_output, atol=1.2e-2, rtol=1e-3)
@pytest.mark.parametrize("model", ["bd2lcco/Qwen3-0.6B-finetuned"])