Skip to content

[CI] fix oom of moe ci #4172

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
gyohuangxin merged 2 commits into from
Jul 10, 2026
Merged

[CI] fix oom of moe ci #4172

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0148af7
fix oom of moe ci
benenzhu Jul 10, 2026
393935b
Merge branch 'main' into fix/moe-2stage-oom
benenzhu Jul 10, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions op_tests/test_moe_2stage.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@

import torch
import itertools
import gc
from contextlib import nullcontext
import aiter
from aiter import dtypes
Expand Down Expand Up @@ -1057,6 +1058,9 @@ def _write_bench_csv(rows):
os.environ.pop("AITER_BF16_FP8_MOE_BOUND", None)
else:
os.environ["AITER_BF16_FP8_MOE_BOUND"] = _old_moe_bound
# Reclaim per-case VRAM to avoid OOM under fragmentation.
gc.collect()
torch.cuda.empty_cache()
if ret is None:
continue
ret.update(extras)
Expand Down
Loading