Skip to content

Pull requests: huggingface/transformers

New pull request New
1,203 Open 23,902 Closed
1,203 Open 23,902 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

Less verbose weight-loading tqdm when stdout is not a TTY (fixes #44303)
#44345 opened Feb 27, 2026 by manavshrivastavagit Loading…
1
Fix tokenizer_class in tokenizer_config.json for Qwen3.5 save_pretrained (fixes #44297)
#44344 opened Feb 27, 2026 by manavshrivastavagit Loading…
1
2
Fix ANSI codes in loading_report when stdout is not a TTY (fixes #44336)
#44343 opened Feb 27, 2026 by manavshrivastavagit Loading…
1
1
[docs]
#44342 opened Feb 27, 2026 by stevhliu Loading…
1
Fix ANSI color handling in loading report for interactive terminals (#44336)
#44341 opened Feb 27, 2026 by Kokonico Loading…
1
3
Fix speculative tests that are flaky with SDPA
#44340 opened Feb 27, 2026 by Rocketknight1 Loading…
2
model: Add DEIMv2 to Transformers
#44339 opened Feb 27, 2026 by harshaljanjani • Draft
5 tasks done
1
1
Update distributed tests
#44338 opened Feb 27, 2026 by SunMarc Loading…
3 of 7 tasks
2
[docs] kernelconfig fix
#44337 opened Feb 27, 2026 by stevhliu Loading…
1
Refactor RoFormer output tracing with @capture_outputs and @can_return_tuple
#44335 opened Feb 27, 2026 by ManasVardhan Loading…
4 tasks done
1
Fix CookieCutter
#44334 opened Feb 27, 2026 by NielsRogge Loading…
1
Refactor ALBERT to use named attributes and remove redundant return_dict=True
#44333 opened Feb 27, 2026 by ManasVardhan Loading…
1
[timesfm2_5] fix timesfm2.5 loss
#44331 opened Feb 27, 2026 by kashif Loading…
5 tasks
2
Remove cache_position in more models
#44330 opened Feb 27, 2026 by Cyrilvallez Loading…
2
Enable Liger Kernel when doing hyperparameter search.
#44329 opened Feb 27, 2026 by linfeng-du Loading…
1 of 5 tasks
1
2
skip 2 invalid test cases for voxtral_realtime model
#44321 opened Feb 27, 2026 by kaixuanliu Loading…
2
Add SAM3-LiteText
#44320 opened Feb 27, 2026 by NielsRogge Loading…
1
2
Support non-gated experts
#44319 opened Feb 27, 2026 by IlyasMoutawwakil Loading…
5 tasks
5
Reduce tqdm verbosity during weight loading
#44316 opened Feb 27, 2026 by jashshah999 Loading…
1
1
add HyperClovaX Vision
#44314 opened Feb 27, 2026 by jp1924 Loading…
2 of 5 tasks
3
docs: Add NeMo Automodel community integration docs
#44304 opened Feb 26, 2026 by adil-a Loading…
3 of 5 tasks
8
Integrate the Neuron device to TrainingArguments
#44302 opened Feb 26, 2026 by michaelbenayoun Loading…
1
Fix: Qwen3 <think> blocks not written during fine-tuning (TRL)
#44301 opened Feb 26, 2026 by likejazz Loading…
2
Dynamic weight conversion applies to VLMs and is recursive
#44300 opened Feb 26, 2026 by zucchini-nlp Loading…
16
Auto detect wrong mapping models
#44298 opened Feb 26, 2026 by ArthurZucker • Draft
3
ProTip! Follow long discussions with comments:>50.