Skip to content

GitLab

Unverified Commit 99b4f080 authored by QiliangCui's avatar QiliangCui Committed by GitHub
Browse files

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


Signed-off-by: default avatarQiliang Cui <derrhein@gmail.com>
parent 020f58ab
Show whitespace changes
Inline Side-by-side
Showing with 2 additions and 3 deletions
tests/entrypoints/llm/test_accuracy.py
View file @ 99b4f080
...@@ -71,9 +71,8 @@ def test_lm_eval_accuracy_v1_engine(model, monkeypatch: pytest.MonkeyPatch): ...@@ -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"
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment