linoytsaban
7530559a04
vace
2025-05-21 19:49:10 +03:00
linoytsaban
b8a371ef69
vace
2025-05-21 19:47:02 +03:00
linoytsaban
f81d4c6a5d
revert
2025-05-21 19:27:43 +03:00
linoytsaban
f91dae9be3
revert
2025-05-21 19:13:21 +03:00
linoytsaban
c665a0412c
revert
2025-05-21 19:11:44 +03:00
linoytsaban
78a8e1e841
vace padding
2025-05-21 18:53:19 +03:00
linoytsaban
5901c2508b
vace padding
2025-05-21 18:50:31 +03:00
linoytsaban
46755bdf1a
vace padding
2025-05-21 18:24:05 +03:00
linoytsaban
c0d22f8757
vace padding
2025-05-21 18:18:45 +03:00
linoytsaban
a66ad3fe88
vace padding
2025-05-21 18:16:35 +03:00
linoytsaban
0279a067f1
vace padding
2025-05-21 18:07:17 +03:00
linoytsaban
b3394d4275
revert
2025-05-21 17:55:32 +03:00
linoytsaban
26dcfd00a4
test
2025-05-21 17:30:32 +03:00
Linoy Tsaban
4a4b05861e
Merge branch 'main' into integrations/wan-vace
2025-05-21 15:45:03 +03:00
Sayak Paul
54af3ca7fd
[chore] allow string device to be passed to randn_tensor. ( #11559 )
...
allow string device to be passed to randn_tensor.
2025-05-21 14:55:54 +05:30
Aryan
23f6bc1201
fix bug with mask
2025-05-21 02:19:30 +02:00
Aryan
694dcf2cfd
refactor
2025-05-21 00:52:15 +02:00
Aryan
4b14ddd80b
add pipeline test
2025-05-20 21:35:30 +02:00
Aryan
ea301df80b
refactor
2025-05-20 21:35:23 +02:00
Sai Shreyas Bhavanasi
ba8dc7dc49
RegionalPrompting: Inherit from Stable Diffusion ( #11525 )
...
* Refactoring Regional Prompting pipeline to use Diffusion Pipeline instead of Stable Diffusion Pipeline
* Apply style fixes
2025-05-20 15:03:16 -04:00
Steven Liu
23a4ff8488
[docs] Remove fast diffusion tutorial ( #11583 )
...
remove tutorial
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
2025-05-20 08:56:12 -07:00
osrm
8705af0914
docs: fix invalid links ( #11505 )
...
* fix invalid link lora.md
* fix invalid link controlnet_sdxl.md
The Hugging Face models page now uses the tags parameter instead of the other parameter for tag-based filtering. Therefore, to simultaneously apply both the "Stable Diffusion XL" and "ControlNet" tags, the following URL should be used: https://huggingface.co/models?tags=stable-diffusion-xl,controlnet
* fix invalid link cosine_dpm.md
"https://github.com/Stability-AI/stable-audio-tool " -> "https://github.com/Stability-AI/stable-audio-tools "
* Update controlnet_sdxl.md
* Update cosine_dpm.md
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com >
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com >
2025-05-20 08:55:41 -07:00
Linoy Tsaban
5d4f723b57
[LoRA] kijai wan lora support for I2V ( #11588 )
...
* testing
* testing
* testing
* testing
* testing
* i2v
* i2v
* device fix
* testing
* fix
* fix
* fix
* fix
* fix
* Apply style fixes
* empty commit
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2025-05-20 17:11:55 +03:00
Aryan
05c8b42b75
LTX 0.9.7-distilled; documentation improvements ( #11571 )
...
* add guidance rescale
* update docs
* support adaptive instance norm filter
* fix custom timesteps support
* add custom timestep example to docs
* add a note about best generation settings being available only in the original repository
* use original org hub ids instead of personal
* make fix-copies
---------
Co-authored-by: Linoy Tsaban <57615435+linoytsaban@users.noreply.github.com >
2025-05-20 02:29:16 +05:30
Aryan
834bfc6448
add conversion script
2025-05-19 21:56:11 +02:00
Aryan
32ab1c969c
fix no split modules
2025-05-19 21:53:18 +02:00
Aryan
f9865ff1ff
make fix-copies
2025-05-19 21:51:16 +02:00
Aryan
d15d090d55
initial support
2025-05-19 21:50:57 +02:00
Quentin Gallouédec
c8bb1ff53e
Use HF Papers ( #11567 )
...
* Use HF Papers
* Apply style fixes
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2025-05-19 06:22:33 -10:00
Dhruv Nair
799adf4a10
[Single File] Fix loading for LTX 0.9.7 transformer ( #11578 )
...
update
2025-05-19 06:22:13 -10:00
Sayak Paul
00f9273da2
[WIP][LoRA] start supporting kijai wan lora. ( #11579 )
...
* start supporting kijai wan lora.
* diff_b keys.
* Apply suggestions from code review
Co-authored-by: Aryan <aryan@huggingface.co >
* merge ready
---------
Co-authored-by: Aryan <aryan@huggingface.co >
2025-05-19 20:47:44 +05:30
Linoy Tsaban
ceb7af277c
[LoRA] support non-diffusers LTX-Video loras ( #11572 )
...
* support non diffusers loras for ltxv
* Update src/diffusers/loaders/lora_conversion_utils.py
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
* Update src/diffusers/loaders/lora_pipeline.py
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
* Apply style fixes
* empty commit
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2025-05-19 12:59:55 +03:00
Sayak Paul
6918f6d19a
[docs] tip for group offloding + quantization ( #11576 )
...
* tip for group offloding + quantization
Co-authored-by: Aryan VS <contact.aryanvs@gmail.com >
* Apply suggestions from code review
Co-authored-by: Aryan <aryan@huggingface.co >
---------
Co-authored-by: Aryan VS <contact.aryanvs@gmail.com >
Co-authored-by: Aryan <aryan@huggingface.co >
2025-05-19 14:49:15 +05:30
apolinário
915c537891
Revert error to warning when loading LoRA from repo with multiple weights ( #11568 )
2025-05-19 13:33:43 +05:30
space_samurai
8270fa58e4
Doc update ( #11531 )
...
Update docs/source/en/using-diffusers/inpaint.md
2025-05-19 13:32:08 +05:30
Yao Matrix
1a10fa0c82
enhance value guard of _device_agnostic_dispatch ( #11553 )
...
enhance value guard
Signed-off-by: Matrix Yao <matrix.yao@intel.com >
2025-05-19 06:05:32 +05:30
Sayak Paul
9836f0e000
[docs] Regional compilation docs ( #11556 )
...
* add regional compilation docs.
* minor.
* reviwer feedback.
* Update docs/source/en/optimization/torch2.0.md
Co-authored-by: Ilyas Moutawwakil <57442720+IlyasMoutawwakil@users.noreply.github.com >
---------
Co-authored-by: Ilyas Moutawwakil <57442720+IlyasMoutawwakil@users.noreply.github.com >
2025-05-15 19:11:24 +05:30
Sayak Paul
20379d9d13
[tests] add tests for combining layerwise upcasting and groupoffloading. ( #11558 )
...
* add tests for combining layerwise upcasting and groupoffloading.
* feedback
2025-05-15 17:16:44 +05:30
Animesh Jain
3a6caba8e4
[gguf] Refactor __torch_function__ to avoid unnecessary computation ( #11551 )
...
* [gguf] Refactor __torch_function__ to avoid unnecessary computation
This helps with torch.compile compilation lantency. Avoiding unnecessary
computation should also lead to a slightly improved eager latency.
* Apply style fixes
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2025-05-15 14:38:18 +05:30
Dhruv Nair
4267d8f4eb
[Single File] GGUF/Single File Support for HiDream ( #11550 )
...
* update
* update
* update
* update
* update
* update
* update
2025-05-15 12:25:18 +05:30
Seokhyeon Jeong
f4fa3beee7
[tests] Add torch.compile test for UNet2DConditionModel ( #11537 )
...
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
2025-05-14 11:26:12 +05:30
Anwesha Chowdhury
7e3353196c
Fix deprecation warnings in test_ltx_image2video.py ( #11538 )
...
Fixed 2 warnings that were raised during running LTXImageToVideoPipelineFastTests
Co-authored-by: achowdhury1211@gmail.com <anwesha@LAPTOP-E5QGFMOQ>
2025-05-13 08:05:52 -10:00
Meatfucker
8c249d1401
Update pipeline_flux_img2img.py to add missing vae_slicing and vae_tiling calls. ( #11545 )
...
* Update pipeline_flux_img2img.py
Adds missing vae_slicing and vae_tiling calls to FluxImage2ImagePipeline
* Update src/diffusers/pipelines/flux/pipeline_flux_img2img.py
Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com >
* Update src/diffusers/pipelines/flux/pipeline_flux_img2img.py
Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com >
* Update src/diffusers/pipelines/flux/pipeline_flux_img2img.py
Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com >
* Update src/diffusers/pipelines/flux/pipeline_flux_img2img.py
Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com >
---------
Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com >
2025-05-13 12:31:45 -04:00
Sayak Paul
b555a03723
[tests] Enable testing for HiDream transformer ( #11478 )
...
* add tests for hidream transformer model.
* fix
* Update tests/models/transformers/test_models_transformer_hidream.py
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com >
---------
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com >
2025-05-13 21:01:20 +05:30
Aryan
06fee551e9
LTX Video 0.9.7 ( #11516 )
...
* add upsampling pipeline
* ltx upsample pipeline conversion; pipeline fixes
* make fix-copies
* remove print
* add vae convenience methods
* update
* add tests
* support denoising strength for upscaling & video-to-video
* update docs
* update doc checkpoints
* update docs
* fix
---------
Co-authored-by: Linoy Tsaban <57615435+linoytsaban@users.noreply.github.com >
2025-05-13 14:57:03 +05:30
Linoy Tsaban
8b99f7e157
[LoRA] small change to support Hunyuan LoRA Loading for FramePack ( #11546 )
...
init
2025-05-13 10:15:06 +03:00
Kenneth Gerald Hamilton
07dd6f8c0e
[train_dreambooth.py] Fix the LR Schedulers when num_train_epochs is passed in a distributed training env ( #11239 )
...
Co-authored-by: Linoy Tsaban <57615435+linoytsaban@users.noreply.github.com >
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
2025-05-13 07:34:01 +05:30
johannaSommer
f8d4a1e283
fix: remove torch_dtype="auto" option from docstrings ( #11513 )
...
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com >
2025-05-12 15:42:35 -10:00
Abdellah Oumida
ddd0cfb497
Fix typo in train_diffusion_orpo_sdxl_lora_wds.py ( #11541 )
2025-05-12 15:28:29 -10:00
Zhong-Yu Li
4f438de35a
Add VisualCloze ( #11377 )
...
* VisualCloze
* style quality
* add docs
* add docs
* typo
* Update docs/source/en/api/pipelines/visualcloze.md
* delete einops
* style quality
* Update src/diffusers/pipelines/visualcloze/pipeline_visualcloze.py
* reorg
* refine doc
* style quality
* typo
* typo
* Update src/diffusers/image_processor.py
* add comment
* test
* style
* Modified based on review
* style
* restore image_processor
* update example url
* style
* fix-copies
* VisualClozeGenerationPipeline
* combine
* tests docs
* remove VisualClozeUpsamplingPipeline
* style
* quality
* test examples
* quality style
* typo
* make fix-copies
* fix test_callback_cfg and test_save_load_dduf in VisualClozePipelineFastTests
* add EXAMPLE_DOC_STRING to VisualClozeGenerationPipeline
* delete maybe_free_model_hooks from pipeline_visualcloze_combined
* Apply suggestions from code review
* fix test_save_load_local test; add reason for skipping cfg test
* more save_load test fixes
* fix tests in generation pipeline tests
2025-05-13 02:46:51 +05:30
