The Wayback Machine - https://web.archive.org/web/20260226052534/https://github.com/huggingface/transformers/pulls
Skip to content

Pull requests: huggingface/transformers

New pull request New
1,198 Open 23,861 Closed
1,198 Open 23,861 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇄1�7 + 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

[docs] optimizers, hyperparam search, training features
#44290 opened Feb 26, 2026 by stevhliu Loading 1�7
1
fix: replace 8 bare except clauses with except Exception
#44289 opened Feb 26, 2026 by haosenwang1018 Loading 1�7
1
Fix TypeError in modeling_rope_utils.py when ignore_keys_at_rope_vali 1�7
#44288 opened Feb 25, 2026 by somAzzz Loading 1�7
1 of 5 tasks
Fix mutable default arguments and resource leaks
#44287 opened Feb 25, 2026 by jashshah999 Loading 1�7
1
Add future model support for Fouroversix
#44286 opened Feb 25, 2026 by kathrynle20 Loading 1�7
Add VidEoMT
#44285 opened Feb 25, 2026 by NielsRogge Loading 1�7
2
Fix Mixtral auxiliary loss computation when output_router_logits=False
#44284 opened Feb 25, 2026 by paipeline Loading 1�7
1
1
Fix special token maps BC
#44281 opened Feb 25, 2026 by ArthurZucker • Draft
1
[prototype] use real tests frm the doc
#44277 opened Feb 25, 2026 by tarekziade • Draft
@tarekziade
3
Add correct typing to custom images_kwargs in ProcessorsKwargs
#44270 opened Feb 25, 2026 by yonigozlan Loading 1�7
2
Add ProcessingKwargs ImagesKwargs etc. to docs
#44269 opened Feb 25, 2026 by yonigozlan Loading 1�7
2
[Moe] Enable aux loss automatically when in training + coef is not 0
#44264 opened Feb 24, 2026 by vasqu • Draft
8
Update common tests Trainer
#44260 opened Feb 24, 2026 by SunMarc Loading 1�7
3
Async data producer
#44259 opened Feb 24, 2026 by winglian • Draft
5 tasks
2
[Ernie 4.5 VL Moe] Change revision
#44258 opened Feb 24, 2026 by vasqu Loading 1�7
2
use nanmean for aggregating loss
#44257 opened Feb 24, 2026 by winglian Loading 1�7
5 tasks
9
Fix CLI NameError: name 'TypeAdapter' is not defined
#44256 opened Feb 24, 2026 by albertvillanova Loading 1�7
1
[vllm + v5 fix] update deepseek v2 tokenizer class for v5
#44255 opened Feb 24, 2026 by itazap Loading 1�7
4
Fix fast tokenizers overwriting custom pre_tokenizer from tokenizer.json
#44254 opened Feb 24, 2026 by mario-sanz Loading 1�7
7
Speed create_import_structure up with os.scandir()
#44253 opened Feb 24, 2026 by Rocketknight1 Loading 1�7
1
1
Timm unification continued
#44252 opened Feb 24, 2026 by zucchini-nlp Loading 1�7
17
Add jina-embeddings-v3 Model
#44251 opened Feb 24, 2026 by Sai-Suraj-27 Loading 1�7
4 of 5 tasks
9
fix zero3 init config
#44236 opened Feb 23, 2026 by SunMarc Loading 1�7
3
chore(qa): Add type checking to src/transformers/generation
#44233 opened Feb 23, 2026 by tarekziade Loading 1�7
@tarekziade
15
[Performance] FP8 Grouped and Batched Matmuls
#44231 opened Feb 23, 2026 by IlyasMoutawwakil • Draft
5 tasks
2
ProTip! Follow long discussions with comments:>50.