Skip to content

Pull requests: huggingface/transformers

New pull request New
1,129 Open 23,516 Closed
1,129 Open 23,516 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Fix DeepSpeed model preparation logic in Trainer class
#43780 opened Feb 5, 2026 by qgallouedec Loading…
2
SwanLab: Add support for id and resume arguments in SwanLabCallback
#43779 opened Feb 5, 2026 by surya10602 Loading…
1 of 5 tasks
1
Mamba-1/-2 init weights in mixer class
#43778 opened Feb 5, 2026 by kevinli573 Loading…
2
Bump dev version
#43777 opened Feb 5, 2026 by qgallouedec Loading…
5 tasks
1
Refactor trainer data_colllator and callbacks tests
#43776 opened Feb 5, 2026 by SunMarc Loading…
1
fix(moe): normalize auxiliary loss by top_k for correct load balancing
#43775 opened Feb 5, 2026 by Mr-Neutr0n Loading…
1
1
Add activation offloading to trainer
#43774 opened Feb 5, 2026 by mbtariq82 Loading…
4 of 5 tasks
1
2
Fix-release-ubild
#43773 opened Feb 5, 2026 by ArthurZucker Loading…
2
[Modular Dependencies] Fixup qwen rms norms
#43772 opened Feb 5, 2026 by vasqu Loading…
6
fix: Add MXFP4 MoE/attention backward kernels
#43771 opened Feb 5, 2026 by leoneperdigao Loading…
4 tasks done
1
2
Remove unconditional train_batch_size assignment
#43770 opened Feb 5, 2026 by lordaarush Loading…
2 of 5 tasks
1
7
Add Voxtral Realtime Audio New model
#43769 opened Feb 5, 2026 by eustlb Loading…
1
Fix init weights in remote code
#43768 opened Feb 5, 2026 by zucchini-nlp Loading…
3
[Model] Add PP-Chart2Table Model Support
#43767 opened Feb 5, 2026 by XingweiDeng Loading…
5 tasks
1
Fix convert_rope_params_to_dict so it uses rope_theta from the config
#43766 opened Feb 5, 2026 by hmellor Loading…
6
[core] Faster and thread-safe check_model_inputs implementation
#43765 opened Feb 5, 2026 by Cyrilvallez Loading…
2
Modify ModernBERT's default attention implementation to stop using FA
#43764 opened Feb 5, 2026 by YangKai0616 Loading…
4
Improved agents
#43763 opened Feb 5, 2026 by tarekziade Loading…
8
Widen match condition for _can_record_outputs
#43762 opened Feb 5, 2026 by molbap Loading…
1
7
Avoid hard failure for gpt-oss GGUF architecture by falling back to g…
#43757 opened Feb 5, 2026 by TheSanjBot Loading…
2 of 5 tasks
1
2
Ernie4 5 vl moe
#43755 opened Feb 5, 2026 by kaixuanliu Loading…
2
Update KERNELS_MIN_VERSION to 0.10.2 to be the same as setup.py
#43753 opened Feb 5, 2026 by cyyever Loading…
1
Param2moe v4.52.3
#43752 opened Feb 5, 2026 by bhargav-patel-29 Loading…
2 of 5 tasks
1
Fix ruff warnings
#43751 opened Feb 5, 2026 by cyyever Loading…
1
enable tp for benchmark
#43750 opened Feb 5, 2026 by sywangyi Loading…
1
ProTip! Add no:assignee to see everything that’s not assigned.