Skip to content

Pull requests: huggingface/transformers

New pull request New
1,095 Open 23,205 Closed
1,095 Open 23,205 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fix: Correct Switch Transformers sparse layer logic and outdated spectrogram comment
#43336 opened Jan 17, 2026 by harshaljanjani Loading…
4 of 5 tasks
1
3
Fix typo: interupted -> interrupted
#43333 opened Jan 17, 2026 by NAM-MAN Loading…
1
Clarify instructions for building documentation locally
#43332 opened Jan 17, 2026 by Edge-Explorer Loading…
1 of 5 tasks
1
docs(yoso): fix and stabilize doctest for YOSO
#43331 opened Jan 17, 2026 by Siddhartha7340 Loading…
4 of 5 tasks
1
fix: _get_num_multimodal_tokens video branch (#43329)
#43330 opened Jan 17, 2026 by floor-licker Loading…
1
1
Fix Mxfp4 dequantize
#43326 opened Jan 16, 2026 by Cyrilvallez Loading…
1
6
More V5 pipeline cleanup
#43325 opened Jan 16, 2026 by Rocketknight1 Loading…
5
fix failed test case for XPU
#43324 opened Jan 16, 2026 by kaixuanliu Loading…
1
[WIP] Attempt at cleaning backbone utils
#43323 opened Jan 16, 2026 by zucchini-nlp Loading…
3
Fix typing hints for different rope parameters per layer type
#43320 opened Jan 16, 2026 by Tcc0403 • Draft
2 of 5 tasks
1
6
43125: Ensure correct checkpoint saving behavior by simplifying Trainer.save_model parallelism logic
#43314 opened Jan 16, 2026 by vaibhav-research Loading…
2 of 5 tasks
1
7
Fix MobileNet v1/v2 image processor default interpolation to BICUBIC
#43313 opened Jan 16, 2026 by VedantMadane Loading…
2
[docs] finetuning
#43312 opened Jan 15, 2026 by stevhliu Loading…
1
[llama4] update rope with new parameter
#43311 opened Jan 15, 2026 by kashif Loading…
5 tasks
3
Replace regex with re
#43310 opened Jan 15, 2026 by Rocketknight1 Loading…
3
[Quantization] Add cutlass kernel for FP8
#43304 opened Jan 15, 2026 by MekkCyber Loading…
17
[DeepSpeed] add weight_mapping to _load_state_dict_into_zero3_model
#43303 opened Jan 15, 2026 by kashif Loading…
5 tasks
1
5
[Feat] Reduces redundant tokenization of <pad> tags to accelerate Qwen3VL.
#43297 opened Jan 15, 2026 by ZLkanyo009 Loading…
6
[docs] pretraining
#43294 opened Jan 14, 2026 by stevhliu Loading…
3
Fix whisper tests
#43291 opened Jan 14, 2026 by eustlb • Draft
3
fix:auto tokenizer pretrainedconfig
#43289 opened Jan 14, 2026 by ansh-info Loading…
2 of 5 tasks
2
[VITS] Add speaking_rate as an optionl forward argument
#43283 opened Jan 14, 2026 by gau-nernst Loading…
5 tasks
2
fix: access rope_theta correctly in Llama4VisionRotaryEmbedding
#43281 opened Jan 14, 2026 by Tcc0403 Loading…
2 of 5 tasks
1
7
Prepare inputs for generation is called from super()
#43280 opened Jan 14, 2026 by zucchini-nlp Loading…
46
[Model] Add PP-OCRV5_server_det Model Support New model
#43274 opened Jan 14, 2026 by XingweiDeng Loading…
5 tasks
1
ProTip! Type g i on any issue or pull request to go back to the issue listing page.