xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-14 06:45:00 +08:00
Code Issues Packages Projects Releases Wiki Activity

[CI Failure] Fix test_kv_cache_model_load_and_run (#27717)

Browse Source 
Signed-off-by: Huamin Li <3ericli@gmail.com>
This commit is contained in:
Huamin Li 2025-10-30 05:27:53 -07:00 committed by GitHub
parent 4464723f22
commit 1994de99ea
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 12 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
tests/quantization/test_fp8.py
View File

@ -49,7 +49,18 @@ def test_model_load_and_run(
KV_CACHE_MODELS = [ KV_CACHE_MODELS = [
# AutoFP8 format using separate .k_scale and .v_scale # AutoFP8 format using separate .k_scale and .v_scale
"nm-testing/Qwen2-1.5B-Instruct-FP8-K-V", # The original checkpoint below was removed from the Hub. To unblock CI and
# until a small replacement with split K/V scales is found, skip this case.
# See PR #27717 for context.
pytest.param(
"nm-testing/Qwen2-1.5B-Instruct-FP8-K-V",
marks=pytest.mark.skip(
reason=(
"Checkpoint removed from HF; temporarily disabling this "
"AutoFP8 split K/V case (PR #27717)."
)
),
),
] ]