xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-19 19:46:02 +08:00
Code Issues Packages Projects Releases Wiki Activity

Renable google/gemma-3-1b-it accuracy test. (#20866)

Browse Source 
Signed-off-by: Qiliang Cui <derrhein@gmail.com>
This commit is contained in:
QiliangCui 2025-07-12 21:48:56 -07:00 committed by GitHub
parent 020f58abcd
commit 99b4f080d8
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
tests/entrypoints/llm/test_accuracy.py
View File

@ -71,9 +71,8 @@ def test_lm_eval_accuracy_v1_engine(model, monkeypatch: pytest.MonkeyPatch):
# Limit compilation time for TPU V1 # Limit compilation time for TPU V1
if model == "google/gemma-3-1b-it": if model == "google/gemma-3-1b-it":
pytest.skip( # TPU + google/gemma-3-1b-it + xet doesn't work well.
"Temporarily disabled due to test failures" m.setenv("HF_HUB_DISABLE_XET", "1")
"(timeout or accuracy mismatch). Re-enable once fixed.")
more_args = "max_model_len=2048,max_num_seqs=64" more_args = "max_model_len=2048,max_num_seqs=64"