xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-24 17:55:50 +08:00
Code Issues Packages Projects Releases Wiki Activity

[CI] Reduce Flakiness For test_spec_decode.py::test_suffix_decoding_acceptance (#30367)

Browse Source 
Signed-off-by: Micah Williamson <micah.williamson@amd.com>
This commit is contained in:
Micah Williamson 2025-12-09 20:35:49 -06:00 committed by GitHub
parent b75f826fca
commit 7d80c73d42
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
tests/v1/e2e/test_spec_decode.py
View File

@ -191,8 +191,8 @@ def test_suffix_decoding_acceptance(
# Expect the acceptance rate to improve.
assert first_accept_rate < last_accept_rate
# Heuristic: expect at least 82.5% acceptance rate at the end.
assert last_accept_rate > 0.825
# Heuristic: expect at least 80.0% acceptance rate at the end.
assert last_accept_rate > 0.80
del spec_llm
torch.cuda.empty_cache()