Skip to content

Pull requests: huggingface/diffusers

New pull request New
294 Open 6,376 Closed
294 Open 6,376 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

Add Newbie Image support
#12803 opened Dec 7, 2025 by Disty0 Loading…
3 tasks
[HunyuanVideo1.5] support step-distilled
#12802 opened Dec 7, 2025 by yiyixuxu Loading…
4
perf: optimize CausalConv3d for wan autoencoders
#12800 opened Dec 6, 2025 by c8ef Loading…
3 of 6 tasks
7
Fix meta tensor error with bitsandbytes quantization and device_map
#12799 opened Dec 6, 2025 by arrdel Loading…
1
3
Improve docstrings and type hints in scheduling_dpmsolver_singlestep.py
#12798 opened Dec 5, 2025 by delmalih Loading…
10 tasks done
personal use
#12795 opened Dec 5, 2025 by naomili0924 Loading…
6 tasks
Z-Image-Turbo ControlNet
#12792 opened Dec 4, 2025 by hlky Loading…
1
4
Add NewbiePipeline and NextDiT_3B_GQA_patch2_Adaln_Refiner_WHIT_CLIP transformer
#12789 opened Dec 4, 2025 by E-Anlia Loading…
4 tasks done
3
Fix duplicate timesteps in DPMSolverMultistepScheduler with sigma conversion methods
#12788 opened Dec 4, 2025 by Aznix07 Loading…
3 tasks done
1
[PRX] Improve model compilation
#12787 opened Dec 3, 2025 by WaterKnight1998 Loading…
4 of 6 tasks
1
6
Disable Sage Attention sm90 backend due to confetti/noisy output
#12785 opened Dec 3, 2025 by arrdel Loading…
1
2
Fix AttributeError in Flux2 dreambooth: rename --validation_image to --validation_image_path
#12784 opened Dec 3, 2025 by arrdel Loading…
1
1
[Wan] Optimize time & memory
#12780 opened Dec 3, 2025 by Fabrice-TIERCELIN Loading…
3 of 6 tasks
2
Flux2: Tensor tuples can cause issues for checkpointing
#12777 opened Dec 2, 2025 by dxqb • Draft
Fixed an issue with abnormal training loss in Dreambooth Flux2.
#12774 opened Dec 2, 2025 by tcaimm Loading…
1 of 6 tasks
[WIP] Rolling KV cache for autoregressive generation
#12773 opened Dec 2, 2025 by zucchini-nlp • Draft
1
1
[WIP] Add Flux2 modular
#12763 opened Dec 1, 2025 by DN6 Loading…
6 tasks
8
from_pipe: Keep existing dtype by default
#12762 opened Dec 1, 2025 by missionfloyd Loading…
2 of 6 tasks
1
1
Add ZImageImg2ImgPipeline
#12751 opened Nov 29, 2025 by CalamitousFelicitousness Loading…
4 of 5 tasks
16
Feature/group offload pinning
#12747 opened Nov 29, 2025 by Aki-07 Loading…
4 of 6 tasks
1
1
Add support for Magcache performance Anything related to performance improvements, profiling and benchmarking
#12744 opened Nov 29, 2025 by AlanPonnachan Loading…
4 of 6 tasks
1
14
[Draft] Flux.1 Dual stream example for NPU
#12727 opened Nov 27, 2025 by zhangtao0408 • Draft
6 tasks
[feat] LongSANA: a minute-length real-time video generation model
#12723 opened Nov 26, 2025 by lawrence-cj Loading…
9
Fix link to diffedit implementation reference
#12708 opened Nov 24, 2025 by JuanFKurucz Loading…
1 of 6 tasks
Feat: fal flashpack support
#12707 opened Nov 24, 2025 by RakshithaKowlikar Loading…
ProTip! Adding no:label will show everything without a label.