transformers
Fix Qwen2.5-Omni / Qwen3-Omni-MoE generation with a compilable cache
#47872

Merged

Fix Qwen2.5-Omni / Qwen3-Omni-MoE generation with a compilable cache #47872

zucchini-nlp merged 5 commits into huggingface:main from jiqing-feng:fix-qwen-omni-static-cache

Fix Qwen2.5-Omni / Qwen3-Omni-MoE generation with a compilable cache

c5a4f584

jiqing-feng marked this pull request as ready for review 57 days ago

Merge branch 'main' into fix-qwen-omni-static-cache

5266b5fb

github-actions requested a review from

ArthurZucker 57 days ago

github-actions requested a review from

Rocketknight1 57 days ago

IlyasMoutawwakil requested a review from

zucchini-nlp 52 days ago

ArthurZucker commented on 2026-08-17

zucchini-nlp commented on 2026-08-17

Compute mrope positions in `generate` instead of overriding masks

a863e6f1

Shorten comments

58d1a7c2

Merge branch 'main' into fix-qwen-omni-static-cache

a35baaf5

IlyasMoutawwakil requested a review from

zucchini-nlp 49 days ago

zucchini-nlp approved these changes on 2026-08-19

zucchini-nlp merged 279dbcf4 into main 48 days ago

Reviewers

zucchini-nlp

ArthurZucker

Rocketknight1

Assignees

No one assigned

Labels

None yet

Milestone

No milestone