Skip to content

Pull requests: huggingface/transformers

New pull request New
1,303 Open 24,886 Closed
1,303 Open 24,886 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

perf(qwen3_vl): replace Conv3d with F.linear in patch embed forward
#45771 opened May 4, 2026 by jashshah999 Contributor Loading…
3 tasks done
1
1
Unwrap text_config in AutoModelFor*.from_config
#45770 opened May 4, 2026 by jamesbraza Contributor Loading…
1
3
feat: add bf16_loss training argument for VRAM-efficient QLoRA
#45769 opened May 3, 2026 by butterwecksolutions Loading…
End-to-end test of Gemma 3 + FA2 construction
#45760 opened May 3, 2026 by jamesbraza Contributor Loading…
1
fix: align attention_mask padding with appended eos token in clvp
#45757 opened May 3, 2026 by CharlieKerfoot Loading…
1
fix attribute access in PermuteForRope._apply
#45756 opened May 3, 2026 by CharlieKerfoot Loading…
fix: return correct forward output in AriaTextForCausalLM
#45755 opened May 3, 2026 by CharlieKerfoot Loading…
1
Fix mps device check for moe histogram routing
#45754 opened May 3, 2026 by belamaran96-coder Loading…
3 of 6 tasks
1
1
Fix unhandled exception noise from background safetensors conversion thread
#45752 opened May 3, 2026 by dhruv7477 Loading…
1
2
Add Conformer model
#45751 opened May 3, 2026 by jonghwanhyeon Contributor Loading…
4 of 6 tasks
1
fix: correct spelling in continuous_api docstring
#45749 opened May 3, 2026 by Dhruv908615 Loading…
6 tasks
1
Optimize Qwen3 RoPE: precompute cos/sin cache for static rope_type
#45748 opened May 2, 2026 by RobTand • Draft
1
Fix split batch size
#45747 opened May 2, 2026 by Prachi-kushwaha Loading…
6 tasks
Fix link to modular transformers documentation
#45746 opened May 2, 2026 by SangbumChoi Contributor Loading…
6 tasks
feat: add crop() to StaticCache layers for assisted generation
#45745 opened May 2, 2026 by ArkaD171717 Loading…
1
3
fix(bitsandbytes): implement reverse_op for Bnb4bitDeserialize and Bnb8bitDeserialize
#45743 opened May 2, 2026 by Kaisan10 Loading…
2 of 3 tasks
5
feat(llama): add has_weight parameter to LlamaRMSNorm for FlashNorm-folded checkpoints (+12.77% e2e on Llama-3.2-1B at bf16/A100)
#45742 opened May 2, 2026 by fm1320 Loading…
5 tasks done
4
deepseek r1 distilled tokenizer fix for qwen2 mapping
#45741 opened May 2, 2026 by itazap Collaborator Loading…
2
Fix IndexError in sdpa_mask and flex_attention_mask for 0D tensors during ONNX export
#45740 opened May 2, 2026 by mitre88 Contributor Loading…
1
DeepSeek OCR specifies an incorrect tokenizer class on the Hub
#45739 opened May 1, 2026 by hmellor Member Loading…
4
fix(musicgen_melody): use DynamicCache instead of EncoderDecoderCache
#45738 opened May 1, 2026 by adityachoksi2512 Loading…
1
4
Add regression test for MusicgenMelody audio conditioning (GH #45647)
#45737 opened May 1, 2026 by voodoovampire Loading…
1
3
fix(quantizers): make user-supplied skip_modules additive with auto-detected defaults
#45734 opened May 1, 2026 by xodn348 Loading…
1
2
[CB] Fixes for SDPA and CPU offloading
#45733 opened May 1, 2026 by remi-or Collaborator Loading…
@remi-or
1
[skills] fine-tuning
#45732 opened Apr 30, 2026 by stevhliu Member • Draft
1
ProTip! Updated in the last three days: updated:>2026-04-30.