Pull Requests huggingface/diffusers

fix autoencoder_kl_wan.AutoencoderKLWan.py tiling bugs

#12336 by oneflyingfish was closed 2025-09-17 02:32

Fix autoencoder_kl_wan.py bugs for Wan2.2 VAE close-to-merge

#12335 by franciszzj was merged 2025-09-16 23:43

Convert alphas for embedders for sd-scripts to ai toolkit conversion

#12332 by rockerBOO was merged 2025-09-18 07:11

[docs] CP

#12331 by stevhliu was merged 2025-09-30 16:33

Add RequestScopedPipeline for safe concurrent inference, tokenizer lock and non-mutating retrieve_timesteps

#12328 by FredyRivera-dev was merged 2025-09-18 06:03

Support ControlNet for Qwen-Image-Edit

#12325 by dimitribarbot was closed 2025-10-13 08:51

Add Wan2.2 VACE - Fun

#12324 by linoytsaban was merged 2025-09-15 16:01

[docs] Attention backends

#12320 by stevhliu was merged 2025-09-23 17:59

use kernels to support `_flash_hub` attention backend

#12318 by ParagEkbote was closed 2025-09-25 12:14

[tests] Single scheduler in lora tests great-unbloating

#12315 by sayakpaul was merged 2025-09-24 03:06

[docs] Prompting

#12312 by stevhliu was merged 2025-10-14 20:53

Use SDP on BF16 in GPU/HPU migration

#12310 by dsocek was merged 2025-09-12 18:00

fix for the qwen controlnet pipeline - wrong device can be used

#12309 by ttio2tech was merged 2025-09-10 18:59

[modular]some small fix

#12307 by yiyixuxu was merged 2025-09-29 21:42

[feat] cache allocator warmup for `from_single_model`

#12305 by sayakpaul was merged 2025-09-10 07:25

[docs] AOBaseConfig

#12302 by stevhliu was merged 2025-09-09 02:24

Support ControlNet-Inpainting for Qwen-Image

#12301 by haofanwang was merged 2025-09-09 00:59

[tests] cache non lora pipeline outputs. great-unbloating

#12298 by sayakpaul was merged 2025-10-01 03:32

adjust criteria for marigold-intrinsics example on XPU

#12290 by kaixuanliu was merged 2025-09-09 21:36

Fix many type hint errors

#12289 by DefTruth was merged 2025-09-17 04:52

[docs] components_to_quantize

#12287 by stevhliu was merged 2025-09-05 01:29

[docs] Pipeline group offloading

#12286 by stevhliu was merged 2025-09-10 02:56

[docs] Distributed inference

#12285 by stevhliu was merged 2025-09-29 18:24

[core] feat: support group offloading at the pipeline level

#12283 by sayakpaul was merged 2025-09-10 03:39

[docs] Inference section cleanup

#12281 by stevhliu was merged 2025-09-05 16:34

[docs] Sharing pipelines/models

#12280 by stevhliu was merged 2025-09-04 18:43

[tests] disable xformer tests for pipelines it isn't popular.

#12277 by sayakpaul was merged 2025-09-24 03:32

[quantization] feat: support aobaseconfig classes in `TorchAOConfig`

#12275 by sayakpaul was merged 2025-09-29 12:34

[modular] fix flux modular pipelines for t2i and i2i

#12272 by sayakpaul was merged 2025-09-10 07:09

Deprecate slicing and tiling methods from `DiffusionPipeline` great-unbloating

#12271 by sayakpaul was merged 2025-09-11 04:34