Skip to content

Pull requests: huggingface/diffusers

New pull request New
269 Open 6,525 Closed
269 Open 6,525 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

Improve docstrings and type hints in scheduling_ddim_cogvideox.py
#12992 opened Jan 17, 2026 by delmalih Loading…
10 tasks done
Fix QwenImageEditPlus max_sequence_length and transformer text seq len kwargs
#12991 opened Jan 17, 2026 by tonera Loading…
3 tasks
Bugfix: Context Parallel crash when seq-len not divisible by mesh size
#12988 opened Jan 16, 2026 by tsu-bin Loading…
[Qwen] avoid creating attention masks when there is no padding
#12987 opened Jan 16, 2026 by kashif Loading…
6 tasks
5
[Type Check] for Qwen-Image-Edit and Wan
#12985 opened Jan 16, 2026 by Jayce-Ping Loading…
3 of 6 tasks
z-image support npu
#12979 opened Jan 15, 2026 by luren55 Loading…
1 of 6 tasks
2
Resnet only use contiguous in training mode.
#12977 opened Jan 15, 2026 by jiqing-feng Loading…
1
9
[wip] switch to transformers main again.
#12976 opened Jan 14, 2026 by sayakpaul • Draft
9
[QwenImageEditPlus] batch size > 1 support
#12968 opened Jan 12, 2026 by kashif Loading…
6 tasks
4
Fix the bug in detecting maximal token length in longcat image pipeline
#12965 opened Jan 11, 2026 by JayceonHo Loading…
6 tasks done
Make convert_ldm_clip_checkpoint more flexible about maxtoken
#12964 opened Jan 10, 2026 by ppbrown Loading…
1
added PhotoDoodle pipeline
#12961 opened Jan 10, 2026 by AMEERAZAM08 Loading…
6 tasks
1
[modular] add auto_docstring & more doc related refactors
#12958 opened Jan 10, 2026 by yiyixuxu Loading…
17
ZImageTransformer2D: Only build attention mask if seqlens are not equal
#12955 opened Jan 10, 2026 by Beinsezii Loading…
1 of 6 tasks
1
LTX2 distilled checkpoint support
#12934 opened Jan 9, 2026 by rootonchair Loading…
6 tasks
1
40
Fibo edit pipeline
#12930 opened Jan 8, 2026 by galbria Loading…
4 of 6 tasks
1
fix(examples): explicitly pass resolution to text-to-image inference
#12929 opened Jan 8, 2026 by hoanhle Loading…
3 of 6 tasks
[feat] Add UCGM Scheduler
#12912 opened Jan 5, 2026 by QAQdev Loading…
2 of 6 tasks
8
Discrete diffusion in diffusers
#12911 opened Jan 4, 2026 by kashif • Draft
6 tasks
1
[group offloading] avoid unnecessary moving out to speed up inference
#12910 opened Jan 4, 2026 by gameofdimension Loading…
6 tasks
1
Split attention backends
#12870 opened Dec 21, 2025 by dxqb • Draft
1
Fix LongCat LoRA load/unload and add regression test
#12867 opened Dec 21, 2025 by surajyadav-research Loading…
6
Upgrade GitHub Actions to latest versions
#12866 opened Dec 20, 2025 by salmanmkc Loading…
1
[tests] QwenImage Layer tests
#12858 opened Dec 18, 2025 by sayakpaul Loading…
2
refactor: replace fa3 wrapper with original fa3 in attention backends registry stale Issues that haven't received updates
#12851 opened Dec 17, 2025 by badayvedat Loading…
1 of 6 tasks
6
ProTip! Adding no:label will show everything without a label.