Skip to content

Pull requests: huggingface/transformers

New pull request New
1,274 Open 24,509 Closed
1,274 Open 24,509 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 UnboundLocalError in invert_attention_mask by adding proper shape…
#45247 opened Apr 5, 2026 by gagandhakrey Loading…
6 tasks
Let's CI go great
#45244 opened Apr 4, 2026 by ydshieh Loading…
4
fix: restore mypy type checking for PreTrainedConfig subclasses (#45071)
#45240 opened Apr 4, 2026 by shhKnight30 Loading…
4 tasks done
1
resize_token_embeddings does not resize lm_head
#45236 opened Apr 4, 2026 by KoichiYasuoka Loading…
2 of 6 tasks
1
feat/rfc/poc: Agnostic GPU
#45235 opened Apr 4, 2026 by kallewoof • Draft
2 of 6 tasks
2
feat: make timesfm2_5 onnx export compatible
#45233 opened Apr 4, 2026 by pdufour Loading…
5 of 6 tasks
1
[docs] static model rules
#45232 opened Apr 3, 2026 by stevhliu Loading…
2
fix: remove nonexistent PILImageResampling import from video_processing_utils
#45227 opened Apr 3, 2026 by akhilc08 Loading…
1
2
fix: handle trailing replacement character in Whisper word timestamp decoding
#45226 opened Apr 3, 2026 by akhilc08 Loading…
2 tasks
1
2
Fix: ObjectDetectionPipeline batch inference only returns first image results
#45223 opened Apr 3, 2026 by Talhax55z Loading…
1 task done
1
fix(gemma3, gemma4): default token_type_ids to zeros for text-only training
#45222 opened Apr 3, 2026 by jashshah999 Loading…
1
1
user friendly error when loading audio from video
#45221 opened Apr 3, 2026 by eustlb Loading…
2
Multimodal serve support
#45220 opened Apr 3, 2026 by SunMarc Loading…
8
reduce memory for gemma4 moe model in tp
#45219 opened Apr 3, 2026 by sywangyi Loading…
1
Proposal: Agent-first CLI
#45218 opened Apr 3, 2026 by LysandreJik • Draft
2
[Qwen3_5]Remove unnecessary masked_fill_ in torch_chunk_gated_delta_rule attention computation: "attn = (q_i @ k_i.transpose(-1, -2) * decay_mask[:, :, i]).masked_fill_(mask, 0)"
#45215 opened Apr 3, 2026 by ENg-122 Loading…
1
cohere_asr: fix bug for model_parallel_beam_search test case
#45214 opened Apr 3, 2026 by kaixuanliu Loading…
1
DO NOT MERGE - model creation skill
#45213 opened Apr 3, 2026 by tarekziade • Draft
6 tasks
@tarekziade
3
musicflamingo: add test support for Intel XPU device
#45212 opened Apr 3, 2026 by kaixuanliu Loading…
1
[Qwen3MoE] Fix wrong return type annotation in Qwen3MoeSparseMoeBlock.forward
#45211 opened Apr 3, 2026 by matdou Loading…
4 of 6 tasks
1
1
nomic_bert: make the test suitable for general device.
#45209 opened Apr 3, 2026 by kaixuanliu Loading…
1
[Gemma4] Add docstrings for Per-Layer Embeddings (PLE) pipeline
#45207 opened Apr 3, 2026 by w4nderlust Loading…
1
1
fix bug for videomt model device mismatch
#45204 opened Apr 3, 2026 by kaixuanliu Loading…
1
fix gemma4 has flash-attention incompatbile head-dim=512
#45202 opened Apr 2, 2026 by Qubitium Loading…
3 of 6 tasks
7
fix(models): Resolve regressions in Wav2Vec2PhonemeCTCTokenizer (wav2vec2-lv-60-espeak-cv-ft)
#45199 opened Apr 2, 2026 by harshaljanjani Loading…
4 of 6 tasks
1
1
ProTip! Type g i on any issue or pull request to go back to the issue listing page.