AI/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-27 17:22:53 +03:00
Code Activity
6,178 Commits 740 Branches 94 Tags
modular-doc
Commit Graph

6178 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Charchit Sharma
ff263947ad Fix rotary positional embedding dimension mismatch in Wan and SkyReels V2 transformers (#12594) 
* Fix rotary positional embedding dimension mismatch in Wan and SkyReels V2 transformers

- Store t_dim, h_dim, w_dim as instance variables in WanRotaryPosEmbed and SkyReelsV2RotaryPosEmbed __init__
- Use stored dimensions in forward() instead of recalculating with different formula
- Fixes inconsistency between init (using // 6) and forward (using // 3)
- Ensures split_sizes matches the dimensions used to create rotary embeddings

* quality fix

---------

Co-authored-by: Charchit Sharma <charchitsharma@A-267.local>
 2025-11-11 11:45:36 -10:00
Dhruv Nair
66e6a0215f [CI] Remove unittest dependency from testing_utils.py (#12621) 
* update

* Update tests/testing_utils.py

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* Update tests/testing_utils.py

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* Apply style fixes

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
 2025-11-11 16:40:39 +05:30
Cesaryuan
5a47442f92 Fix: update type hints for Tuple parameters across multiple files to support variable-length tuples (#12544) 
* Fix: update type hints for Tuple parameters across multiple files to support variable-length tuples

* Apply style fixes

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
 2025-11-10 13:57:52 -08:00
Dhruv Nair
8f6328c4a4 [Modular] Clean up docs (#12604) 
update

Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2025-11-10 23:37:29 +05:30
Dhruv Nair
8d45f219d0 Fix Context Parallel validation checks (#12446) 
* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2025-11-10 23:37:07 +05:30
Yashwant Bezawada
0fd58c7706 fix: correct import path for load_model_dict_into_meta in conversion scripts (#12616) 
The function load_model_dict_into_meta was moved from modeling_utils.py to
model_loading_utils.py but the imports in the conversion scripts were not
updated, causing ImportError when running these scripts.

This fixes the import in 6 conversion scripts:
- scripts/convert_sd3_to_diffusers.py
- scripts/convert_stable_cascade_lite.py
- scripts/convert_stable_cascade.py
- scripts/convert_stable_audio.py
- scripts/convert_sana_to_diffusers.py
- scripts/convert_sana_controlnet_to_diffusers.py

Fixes #12606
 2025-11-10 14:47:18 +05:30
Dhruv Nair
35d703310c [CI] Fix typo in uv install (#12618) 
update
 2025-11-10 13:22:46 +05:30
YiYi Xu
b455dc94a2 [modular] wan! (#12611) 
* update, remove intermediaate_inputs

* support image2video

* revert dynamic steps to simplify

* refactor vae encoder block

* support flf2video!

* add support for wan2.2 14B

* style

* Apply suggestions from code review

* input dynamic step -> additiional input step

* up

* fix init

* update dtype
 2025-11-09 21:48:50 -10:00
Jay Wu
04f9d2bf3d add ChronoEdit (#12593) 
* add ChronoEdit

* add ref to  original function & remove wan2.2 logics

* Update src/diffusers/pipelines/chronoedit/pipeline_chronoedit.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/chronoedit/pipeline_chronoedit.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* add ChronoeEdit test

* add docs

* add docs

* make fix-copies

* fix chronoedit test

---------

Co-authored-by: wjay <wjay@nvidia.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2025-11-09 22:07:00 -08:00
Dhruv Nair
bc8fd864eb [CI] Push test fix (#12617) 
update
 2025-11-10 09:26:14 +05:30
Wang, Yi
a9cb08af39 fix the crash in Wan-AI/Wan2.2-TI2V-5B-Diffusers if CP is enabled (#12562) 
* fix the crash in Wan-AI/Wan2.2-TI2V-5B-Diffusers if CP is enabled

Signed-off-by: Wang, Yi <yi.a.wang@intel.com>

* address review comment

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>

* refine

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>

---------

Signed-off-by: Wang, Yi <yi.a.wang@intel.com>
Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>
 2025-11-07 20:00:13 +05:30
DefTruth
9f669e7b5d feat: enable attention dispatch for huanyuan video (#12591) 
* feat: enable attention dispatch for huanyuan video

* feat: enable attention dispatch for huanyuan video
 2025-11-07 11:22:41 +05:30
Dhruv Nair
8ac17cd2cb [Modular] Some clean up for Modular tests (#12579) 
* update

* update

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2025-11-07 08:19:15 +05:30
Mohammad Sadegh Salehi
e4393fa613 Fix overflow and dtype handling in rgblike_to_depthmap (NumPy + PyTorch) (#12546) 
* Fix overflow in rgblike_to_depthmap by safe dtype casting (torch & NumPy)

* Fix: store original dtype and cast back after safe computation

* Apply style fixes

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
 2025-11-06 08:18:21 -10:00
Junsong Chen
b3e9dfced7 [SANA-Video] Adding 5s pre-trained 480p SANA-Video inference (#12584) 
* 1. add `SanaVideoTransformer3DModel` in transformer_sana_video.py
2. add `SanaVideoPipeline` in pipeline_sana_video.py
3. add all code we need for import `SanaVideoPipeline`

* add a sample about how to use sana-video;

* code update;

* update hf model path;

* update code;

* sana-video can run now;

* 1. add aspect ratio in sana-video-pipeline;
2. add reshape function in sana-video-processor;
3. fix convert pth to safetensor bugs;

* default to use `use_resolution_binning`;

* make style;

* remove unused code;

* Update src/diffusers/models/transformers/transformer_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update src/diffusers/models/transformers/transformer_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update src/diffusers/models/transformers/transformer_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update src/diffusers/pipelines/sana/pipeline_sana_video.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update src/diffusers/models/transformers/transformer_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update src/diffusers/models/transformers/transformer_sana_video.py

* Update src/diffusers/pipelines/sana/pipeline_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update src/diffusers/models/transformers/transformer_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update src/diffusers/pipelines/sana/pipeline_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* support `dispatch_attention_fn`

* 1. add sana-video markdown;
2. fix typos;

* add two test case for sana-video (need check)

* fix text-encoder in test-sana-video;

* Update tests/pipelines/sana/test_sana_video.py

* Update tests/pipelines/sana/test_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update tests/pipelines/sana/test_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update tests/pipelines/sana/test_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update tests/pipelines/sana/test_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update tests/pipelines/sana/test_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update src/diffusers/pipelines/sana/pipeline_sana_video.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* Update src/diffusers/video_processor.py

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>

* make style
make quality
make fix-copies

* toctree yaml update;

* add sana-video-transformer3d markdown;

* Apply style fixes

---------

Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
 2025-11-05 21:08:47 -08:00
Joseph Turian
58f3771545 Add optional precision-preserving preprocessing for examples/unconditional_image_generation/train_unconditional.py (#12596) 
* Add optional precision-preserving preprocessing

* Document decoder caveat for precision flag

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2025-11-06 09:37:31 +05:30
Dhruv Nair
6198f8a12b [Modular] Allow ModularPipeline to load from revisions (#12592) 
* update

* update

* update

* update

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2025-11-06 07:54:24 +05:30
Linoy Tsaban
dcfb18a2d3 [LoRA] add support for more Qwen LoRAs (#12581) 
* fix bug when offload and cache_latents both enabled

* fix
 2025-11-04 14:27:25 +02:00
Sayak Paul
ac5a1e28fc [docs] sort doc (#12586) 
sort doc
 2025-11-04 10:26:07 +05:30
Lev Novitskiy
325a95051b Kandinsky 5.0 Docs fixes (#12582) 
* add transformer pipeline first version

* updates

* fix 5sec generation

* rewrite Kandinsky5T2VPipeline to diffusers style

* add multiprompt support

* remove prints in pipeline

* add nabla attention

* Wrap Transformer in Diffusers style

* fix license

* fix prompt type

* add gradient checkpointing and peft support

* add usage example

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>

* remove unused imports

* add 10 second models support

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* remove no_grad and simplified prompt paddings

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* moved template to __init__

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* moved sdps inside processor

* remove oneline function

* remove reset_dtype methods

* Transformer: move all methods to forward

* separated prompt encoding

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* refactoring

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* refactoring acording to acabbc0033

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* fixed

* style +copies

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: Charles <charles@huggingface.co>

* more

* Apply suggestions from code review

* add lora loader doc

* add compiled Nabla Attention

* all needed changes for 10 sec models are added!

* add docs

* Apply style fixes

* update docs

* add kandinsky5 to toctree

* add tests

* fix tests

* Apply style fixes

* update tests

* minor docs refactoring

* refactor Kandinsky 5.0 Vide docs

* Update docs/source/en/_toctree.yml

---------

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: Charles <charles@huggingface.co>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
 2025-11-03 14:38:07 -10:00
Wang, Yi
1ec28a2c77 ulysses enabling in native attention path (#12563) 
* ulysses enabling in native attention path

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>

* address review comment

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>

* add supports_context_parallel for native attention

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>

* update templated attention

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>

---------

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2025-11-03 11:48:20 -10:00
YiYi Xu
de6173c683 [modular]pass hub_kwargs to load_config (#12577) 
pass hub_kwargs to load_config
 2025-11-03 09:44:42 -10:00
Sayak Paul
8f80dda193 [tests] add tests for flux modular (t2i, i2i, kontext) (#12566) 
* start flux modular tests.

* up

* add kontext

* up

* up

* up

* Update src/diffusers/modular_pipelines/flux/denoise.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* up

* up

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2025-11-02 10:51:11 +05:30
YiYi Xu
cdbf0ad883 [modular] better warn message (#12573) 
better warn message
 2025-11-01 18:45:09 -10:00
Dhruv Nair
5e8415a311 Fix custom code loading in Automodel (#12571) 
update
 2025-11-01 17:04:31 -10:00
Friedrich Schöller
051c8a1c0f Fix Stable Diffusion 3.x pooled prompt embedding with multiple images (#12306) 2025-10-31 10:25:13 -10:00
Dhruv Nair
d54622c267 [Modular] Allow custom blocks to be saved to local_dir (#12381) 
update

Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2025-10-31 13:47:02 +05:30
Dhruv Nair
df8dd77817 [Modular] Fix for custom block kwargs (#12561) 
update
 2025-10-31 00:14:24 +05:30
Pavle Padjin
9f3c0fdcd8 Avoiding graph break by changing the way we infer dtype in vae.decoder (#12512) 
* Changing the way we infer dtype to avoid force evaluation of lazy tensors

* changing way to infer dtype to ensure type consistency

* more robust infering of dtype

* removing the upscale dtype entirely
 2025-10-30 08:39:40 +05:30
galbria
84e16575e4 Bria fibo (#12545) 
* Bria FIBO pipeline

* style fixs

* fix CR

* Refactor BriaFibo classes and update pipeline parameters

- Updated BriaFiboAttnProcessor and BriaFiboAttention classes to reflect changes from Flux equivalents.
- Modified the _unpack_latents method in BriaFiboPipeline to improve clarity.
- Increased the default max_sequence_length to 3000 and added a new optional parameter do_patching.
- Cleaned up test_pipeline_bria_fibo.py by removing unused imports and skipping unsupported tests.

* edit the docs of FIBO

* Remove unused BriaFibo imports and update CPU offload method in BriaFiboPipeline

* Refactor FIBO classes to BriaFibo naming convention

- Updated class names from FIBO to BriaFibo for consistency across the module.
- Modified instances of FIBOEmbedND, FIBOTimesteps, TextProjection, and TimestepProjEmbeddings to reflect the new naming.
- Ensured all references in the BriaFiboTransformer2DModel are updated accordingly.

* Add BriaFiboTransformer2DModel import to transformers module

* Remove unused BriaFibo imports from modular pipelines and add BriaFiboTransformer2DModel and BriaFiboPipeline classes to dummy objects for enhanced compatibility with torch and transformers.

* Update BriaFibo classes with copied documentation and fix import typo in pipeline module

- Added documentation comments indicating the source of copied code in BriaFiboTransformerBlock and _pack_latents methods.
- Corrected the import statement for BriaFiboPipeline in the pipelines module.

* Remove unused BriaFibo imports from __init__.py to streamline modular pipelines.

* Refactor documentation comments in BriaFibo classes to indicate inspiration from existing implementations

- Updated comments in BriaFiboAttnProcessor, BriaFiboAttention, and BriaFiboPipeline to reflect that the code is inspired by other modules rather than copied.
- Enhanced clarity on the origins of the methods to maintain proper attribution.

* change Inspired by to Based on

* add reference link and fix trailing whitespace

* Add BriaFiboTransformer2DModel documentation and update comments in BriaFibo classes

- Introduced a new documentation file for BriaFiboTransformer2DModel.
- Updated comments in BriaFiboAttnProcessor, BriaFiboAttention, and BriaFiboPipeline to clarify the origins of the code, indicating copied sources for better attribution.

---------

Co-authored-by: sayakpaul <spsayakpaul@gmail.com>
 2025-10-28 16:27:48 +05:30
Sayak Paul
55d49d4379 [ci] don't run sana layerwise casting tests in CI. (#12551) 
* don't run sana layerwise casting tests in CI.

* up
 2025-10-28 13:29:51 +05:30
Meatfucker
40528e9ae7 Fix typos in kandinsky5 docs (#12552) 
Update kandinsky5.md

Fix typos
 2025-10-28 02:54:24 -03:00
Wang, Yi
dc622a95d0 fix crash if tiling mode is enabled (#12521) 
* fix crash in tiling mode is enabled

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>

* fmt

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>

---------

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2025-10-27 17:59:20 -10:00
Dhruv Nair
ecfbc8f952 [Pipelines] Enable Wan VACE to run since single transformer (#12428) 
* update

* update

* update

* update

* update
 2025-10-28 09:21:31 +05:30
Sayak Paul
df0e2a4f2c support latest few-step wan LoRA. (#12541) 
* support latest few-step wan LoRA.

* up

* up
 2025-10-28 08:55:24 +05:30
G.O.D
303efd2b8d Improve pos embed for Flux.1 inference on Ascend NPU (#12534) 
improve pos embed for ascend npu

Co-authored-by: felix01.yu <felix01.yu@vipshop.com>
 2025-10-27 16:55:36 -10:00
Lev Novitskiy
5afbcce176 Kandinsky 5 10 sec (NABLA suport) (#12520) 
* add transformer pipeline first version

* updates

* fix 5sec generation

* rewrite Kandinsky5T2VPipeline to diffusers style

* add multiprompt support

* remove prints in pipeline

* add nabla attention

* Wrap Transformer in Diffusers style

* fix license

* fix prompt type

* add gradient checkpointing and peft support

* add usage example

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>

* remove unused imports

* add 10 second models support

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* remove no_grad and simplified prompt paddings

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* moved template to __init__

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* moved sdps inside processor

* remove oneline function

* remove reset_dtype methods

* Transformer: move all methods to forward

* separated prompt encoding

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* refactoring

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* refactoring acording to acabbc0033

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* fixed

* style +copies

* Update src/diffusers/models/transformers/transformer_kandinsky.py

Co-authored-by: Charles <charles@huggingface.co>

* more

* Apply suggestions from code review

* add lora loader doc

* add compiled Nabla Attention

* all needed changes for 10 sec models are added!

* add docs

* Apply style fixes

* update docs

* add kandinsky5 to toctree

* add tests

* fix tests

* Apply style fixes

* update tests

---------

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: Charles <charles@huggingface.co>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
 2025-10-28 07:47:18 +05:30
alirezafarashah
6d1a648602 Fix small inconsistency in output dimension of "_get_t5_prompt_embeds" function in sd3 pipeline (#12531) 
* Fix small inconsistency in output dimension of t5 embeds when text_encoder_3 is None

* first commit

---------

Co-authored-by: Alireza Farashah <alireza.farashah@cn-g017.server.mila.quebec>
Co-authored-by: Alireza Farashah <alireza.farashah@login-2.server.mila.quebec>
 2025-10-27 07:16:43 -10:00
Mikko Lauri
250f5cb53d Add AITER attention backend (#12549) 
* add aiter attention backend

* Apply style fixes

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
 2025-10-27 20:25:02 +05:30
josephrocca
dc6bd1511a Fix Chroma attention padding order and update docs to use lodestones/Chroma1-HD (#12508) 
* [Fix] Move attention mask padding after T5 embedding

* [Fix] Move attention mask padding after T5 embedding

* Clean up whitespace in pipeline_chroma.py

Removed unnecessary blank lines for cleaner code.

* Fix

* Fix

* Update model to final Chroma1-HD checkpoint

* Update to Chroma1-HD

* Update model to Chroma1-HD

* Update model to Chroma1-HD

* Update Chroma model links to Chroma1-HD

* Add comment about padding/masking

* Fix checkpoint/repo references

* Apply style fixes

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
 2025-10-27 16:25:20 +05:30
Sayak Paul
500b9cf184 [chore] Move guiders experimental warning (#12543) 
* move guiders experimental warning to init.

* up
 2025-10-26 07:41:23 -10:00
Dhruv Nair
d34b18c783 Deprecate Stable Cascade (#12537) 
* update

* update

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2025-10-24 22:06:31 +05:30
kaixuanliu
7536f647e4 Loose the criteria tolerance appropriately for Intel XPU devices (#12460) 
* Loose the criteria tolerance appropriately for Intel XPU devices

Signed-off-by: Liu, Kaixuan <kaixuan.liu@intel.com>

* change back the atol value

Signed-off-by: Liu, Kaixuan <kaixuan.liu@intel.com>

* use expectations

Signed-off-by: Liu, Kaixuan <kaixuan.liu@intel.com>

* Update tests/pipelines/kandinsky2_2/test_kandinsky_controlnet.py

---------

Signed-off-by: Liu, Kaixuan <kaixuan.liu@intel.com>
Co-authored-by: Ilyas Moutawwakil <57442720+IlyasMoutawwakil@users.noreply.github.com>
 2025-10-24 12:18:15 +02:00
YiYi Xu
a138d71ec1 HunyuanImage21 (#12333) 
* add hunyuanimage2.1


---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2025-10-23 22:31:12 -10:00
Sayak Paul
bc4039886d fix constants.py to user upper() (#12479) 2025-10-24 12:00:02 +05:30
Dhruv Nair
9c3b58dcf1 Handle deprecated transformer classes (#12517) 
* update

* update

* update
 2025-10-23 16:22:07 +05:30
Aishwarya Badlani
74b5fed434 Fix MPS compatibility in get_1d_sincos_pos_embed_from_grid #12432 (#12449) 
* Fix MPS compatibility in get_1d_sincos_pos_embed_from_grid #12432

* Fix trailing whitespace in docstring

* Apply style fixes

---------

Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
 2025-10-23 16:18:07 +05:30
kaixuanliu
85eb505672 fix CI bug for kandinsky3_img2img case (#12474) 
* fix CI bug for kandinsky3_img2img case

Signed-off-by: Liu, Kaixuan <kaixuan.liu@intel.com>

* update code

Signed-off-by: Liu, Kaixuan <kaixuan.liu@intel.com>

---------

Signed-off-by: Liu, Kaixuan <kaixuan.liu@intel.com>
 2025-10-23 16:17:22 +05:30
Sayak Paul
ccdd96ca52 [tests] Test attention backends (#12388) 
* add a lightweight test suite for attention backends.

* up

* up

* Apply suggestions from code review

* formatting
 2025-10-23 15:09:41 +05:30
Sayak Paul
4c723d8ec3 [CI] xfail the test_wuerstchen_prior test (#12530) 
xfail the test_wuerstchen_prior test
 2025-10-22 08:45:47 -10:00