Skip to content
Merged
Changes from all commits
Commits
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
5 changes: 4 additions & 1 deletion examples/models/qwen3_5_moe/test_chunked_prefill.py
Original file line number Diff line number Diff line change
Expand Up @@ -113,7 +113,10 @@ def test_chunked_prefill_matches_unchunked(self):
logits_chunk.to(torch.float32),
logits_full.to(torch.float32),
rtol=1e-2,
atol=1e-2,
# cross-hardware MLX accumulation noise on near-zero logits (the
# greedy-token assert above is the functional guard; a real chunk-
# boundary desync would diverge by orders of magnitude more)
atol=3e-2,
)


Expand Down
Loading