* Improve incorrect LoRA format error message
* Add flag in PeftLoraLoaderMixinTests to disable text encoder LoRA tests
* Apply changes to LTX2LoraTests
* Further improve incorrect LoRA format error msg following review
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
* Fix QwenImage txt_seq_lens handling
* formatting
* formatting
* remove txt_seq_lens and use bool mask
* use compute_text_seq_len_from_mask
* add seq_lens to dispatch_attention_fn
* use joint_seq_lens
* remove unused index_block
* WIP: Remove seq_lens parameter and use mask-based approach
- Remove seq_lens parameter from dispatch_attention_fn
- Update varlen backends to extract seqlens from masks
- Update QwenImage to pass 2D joint_attention_mask
- Fix native backend to handle 2D boolean masks
- Fix sage_varlen seqlens_q to match seqlens_k for self-attention
Note: sage_varlen still producing black images, needs further investigation
* fix formatting
* undo sage changes
* xformers support
* hub fix
* fix torch compile issues
* fix tests
* use _prepare_attn_mask_native
* proper deprecation notice
* add deprecate to txt_seq_lens
* Update src/diffusers/models/transformers/transformer_qwenimage.py
Co-authored-by: YiYi Xu <yixu310@gmail.com>
* Update src/diffusers/models/transformers/transformer_qwenimage.py
Co-authored-by: YiYi Xu <yixu310@gmail.com>
* Only create the mask if there's actual padding
* fix order of docstrings
* Adds performance benchmarks and optimization details for QwenImage
Enhances documentation with comprehensive performance insights for QwenImage pipeline:
* rope_text_seq_len = text_seq_len
* rename to max_txt_seq_len
* removed deprecated args
* undo unrelated change
* Updates QwenImage performance documentation
Removes detailed attention backend benchmarks and simplifies torch.compile performance description
Focuses on key performance improvement with torch.compile, highlighting the specific speedup from 4.70s to 1.93s on an A100 GPU
Streamlines the documentation to provide more concise and actionable performance insights
* Updates deprecation warnings for txt_seq_lens parameter
Extends deprecation timeline for txt_seq_lens from version 0.37.0 to 0.39.0 across multiple Qwen image-related models
Adds a new unit test to verify the deprecation warning behavior for the txt_seq_lens parameter
* fix compile
* formatting
* fix compile tests
* rename helper
* remove duplicate
* smaller values
* removed
* use torch.cond for torch compile
* Construct joint attention mask once
* test different backends
* construct joint attention mask once to avoid reconstructing in every block
* Update src/diffusers/models/attention_dispatch.py
Co-authored-by: YiYi Xu <yixu310@gmail.com>
* formatting
* raising an error from the EditPlus pipeline when batch_size > 1
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: cdutr <dutra_carlos@hotmail.com>
* Basic implementation of request scheduling
* Basic editing in SD and Flux Pipelines
* Small Fix
* Fix
* Update for more pipelines
* Add examples/server-async
* Add examples/server-async
* Updated RequestScopedPipeline to handle a single tokenizer lock to avoid race conditions
* Fix
* Fix _TokenizerLockWrapper
* Fix _TokenizerLockWrapper
* Delete _TokenizerLockWrapper
* Fix tokenizer
* Update examples/server-async
* Fix server-async
* Optimizations in examples/server-async
* We keep the implementation simple in examples/server-async
* Update examples/server-async/README.md
* Update examples/server-async/README.md for changes to tokenizer locks and backward-compatible retrieve_timesteps
* The changes to the diffusers core have been undone and all logic is being moved to exmaples/server-async
* Update examples/server-async/utils/*
* Fix BaseAsyncScheduler
* Rollback in the core of the diffusers
* Update examples/server-async/README.md
* Complete rollback of diffusers core files
* Simple implementation of an asynchronous server compatible with SD3-3.5 and Flux Pipelines
* Update examples/server-async/README.md
* Fixed import errors in 'examples/server-async/serverasync.py'
* Flux Pipeline Discard
* Update examples/server-async/README.md
* Apply style fixes
* Add thread-safe wrappers for components in pipeline
Refactor requestscopedpipeline.py to add thread-safe wrappers for tokenizer, VAE, and image processor. Introduce locking mechanisms to ensure thread safety during concurrent access.
* Add wrappers.py
* Apply style fixes
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Store vae.config.scaling_factor to prevent missing attr reference
In sdxl advanced dreambooth training script
vae.config.scaling_factor becomes inaccessible after: del vae
when: --cache_latents, and no --validation_prompt
Co-authored-by: Teriks <Teriks@users.noreply.github.com>
Co-authored-by: Linoy Tsaban <57615435+linoytsaban@users.noreply.github.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
* 3 files
* add conditoinal pipeline
* refactor qwen modular
* add layered
* up up
* u p
* add to import
* more refacotr, make layer work
* clean up a bit git add src
* more
* style
* style
* Adding torchao version guard for floatx usage
Summary: TorchAO removing floatx support, added version guard in quantization_config.py
* Adding torchao version guard for floatx usage
Summary: TorchAO removing floatx support, added version guard in quantization_config.py
Altered tests in test_torchao.py to version guard floatx
Created new test to verify version guard of floatx support
* Adding torchao version guard for floatx usage
Summary: TorchAO removing floatx support, added version guard in quantization_config.py
Altered tests in test_torchao.py to version guard floatx
Created new test to verify version guard of floatx support
* Adding torchao version guard for floatx usage
Summary: TorchAO removing floatx support, added version guard in quantization_config.py
Altered tests in test_torchao.py to version guard floatx
Created new test to verify version guard of floatx support
* Adding torchao version guard for floatx usage
Summary: TorchAO removing floatx support, added version guard in quantization_config.py
Altered tests in test_torchao.py to version guard floatx
Created new test to verify version guard of floatx support
* Adding torchao version guard for floatx usage
Summary: TorchAO removing floatx support, added version guard in quantization_config.py
Altered tests in test_torchao.py to version guard floatx
Created new test to verify version guard of floatx support
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
* Initial LTX 2.0 transformer implementation
* Add tests for LTX 2 transformer model
* Get LTX 2 transformer tests working
* Rename LTX 2 compile test class to have LTX2
* Remove RoPE debug print statements
* Get LTX 2 transformer compile tests passing
* Fix LTX 2 transformer shape errors
* Initial script to convert LTX 2 transformer to diffusers
* Add more LTX 2 transformer audio arguments
* Allow LTX 2 transformer to be loaded from local path for conversion
* Improve dummy inputs and add test for LTX 2 transformer consistency
* Fix LTX 2 transformer bugs so consistency test passes
* Initial implementation of LTX 2.0 video VAE
* Explicitly specify temporal and spatial VAE scale factors when converting
* Add initial LTX 2.0 video VAE tests
* Add initial LTX 2.0 video VAE tests (part 2)
* Get diffusers implementation on par with official LTX 2.0 video VAE implementation
* Initial LTX 2.0 vocoder implementation
* Use RMSNorm implementation closer to original for LTX 2.0 video VAE
* start audio decoder.
* init registration.
* up
* simplify and clean up
* up
* Initial LTX 2.0 text encoder implementation
* Rough initial LTX 2.0 pipeline implementation
* up
* up
* up
* up
* Add imports for LTX 2.0 Audio VAE
* Conversion script for LTX 2.0 Audio VAE Decoder
* Add Audio VAE logic to T2V pipeline
* Duplicate scheduler for audio latents
* Support num_videos_per_prompt for prompt embeddings
* LTX 2.0 scheduler and full pipeline conversion
* Add script to test full LTX2Pipeline T2V inference
* Fix pipeline return bugs
* Add LTX 2 text encoder and vocoder to ltx2 subdirectory __init__
* Fix more bugs in LTX2Pipeline.__call__
* Improve CPU offload support
* Fix pipeline audio VAE decoding dtype bug
* Fix video shape error in full pipeline test script
* Get LTX 2 T2V pipeline to produce reasonable outputs
* Make LTX 2.0 scheduler more consistent with original code
* Fix typo when applying scheduler fix in T2V inference script
* Refactor Audio VAE to be simpler and remove helpers (#7)
* remove resolve causality axes stuff.
* remove a bunch of helpers.
* remove adjust output shape helper.
* remove the use of audiolatentshape.
* move normalization and patchify out of pipeline.
* fix
* up
* up
* Remove unpatchify and patchify ops before audio latents denormalization (#9)
---------
Co-authored-by: dg845 <58458699+dg845@users.noreply.github.com>
* Add support for I2V (#8)
* start i2v.
* up
* up
* up
* up
* up
* remove uniform strategy code.
* remove unneeded code.
* Denormalize audio latents in I2V pipeline (analogous to T2V change) (#11)
* test i2v.
* Move Video and Audio Text Encoder Connectors to Transformer (#12)
* Denormalize audio latents in I2V pipeline (analogous to T2V change)
* Initial refactor to put video and audio text encoder connectors in transformer
* Get LTX 2 transformer tests working after connector refactor
* precompute run_connectors,.
* fixes
* Address review comments
* Calculate RoPE double precisions freqs using torch instead of np
* Further simplify LTX 2 RoPE freq calc
* Make connectors a separate module (#18)
* remove text_encoder.py
* address yiyi's comments.
* up
* up
* up
* up
---------
Co-authored-by: sayakpaul <spsayakpaul@gmail.com>
* up (#19)
* address initial feedback from lightricks team (#16)
* cross_attn_timestep_scale_multiplier to 1000
* implement split rope type.
* up
* propagate rope_type to rope embed classes as well.
* up
* When using split RoPE, make sure that the output dtype is same as input dtype
* Fix apply split RoPE shape error when reshaping x to 4D
* Add export_utils file for exporting LTX 2.0 videos with audio
* Tests for T2V and I2V (#6)
* add ltx2 pipeline tests.
* up
* up
* up
* up
* remove content
* style
* Denormalize audio latents in I2V pipeline (analogous to T2V change)
* Initial refactor to put video and audio text encoder connectors in transformer
* Get LTX 2 transformer tests working after connector refactor
* up
* up
* i2v tests.
* up
* Address review comments
* Calculate RoPE double precisions freqs using torch instead of np
* Further simplify LTX 2 RoPE freq calc
* revert unneded changes.
* up
* up
* update to split style rope.
* up
---------
Co-authored-by: Daniel Gu <dgu8957@gmail.com>
* up
* use export util funcs.
* Point original checkpoint to LTX 2.0 official checkpoint
* Allow the I2V pipeline to accept image URLs
* make style and make quality
* remove function map.
* remove args.
* update docs.
* update doc entries.
* disable ltx2_consistency test
* Simplify LTX 2 RoPE forward by removing coords is None logic
* make style and make quality
* Support LTX 2.0 audio VAE encoder
* Apply suggestions from code review
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
* Remove print statement in audio VAE
* up
* Fix bug when calculating audio RoPE coords
* Ltx 2 latent upsample pipeline (#12922)
* Initial implementation of LTX 2.0 latent upsampling pipeline
* Add new LTX 2.0 spatial latent upsampler logic
* Add test script for LTX 2.0 latent upsampling
* Add option to enable VAE tiling in upsampling test script
* Get latent upsampler working with video latents
* Fix typo in BlurDownsample
* Add latent upsample pipeline docstring and example
* Remove deprecated pipeline VAE slicing/tiling methods
* make style and make quality
* When returning latents, return unpacked and denormalized latents for T2V and I2V
* Add model_cpu_offload_seq for latent upsampling pipeline
---------
Co-authored-by: Daniel Gu <dgu8957@gmail.com>
* Fix latent upsampler filename in LTX 2 conversion script
* Add latent upsample pipeline to LTX 2 docs
* Add dummy objects for LTX 2 latent upsample pipeline
* Set default FPS to official LTX 2 ckpt default of 24.0
* Set default CFG scale to official LTX 2 ckpt default of 4.0
* Update LTX 2 pipeline example docstrings
* make style and make quality
* Remove LTX 2 test scripts
* Fix LTX 2 upsample pipeline example docstring
* Add logic to convert and save a LTX 2 upsampling pipeline
* Document LTX2VideoTransformer3DModel forward pass
---------
Co-authored-by: sayakpaul <spsayakpaul@gmail.com>
* docs: add comprehensive docstrings and refine type hints for EDM scheduler methods and config parameters.
* refactor: Add type hints to DPM-Solver scheduler methods.
* feat: Add transformer cache context for conditional and unconditional predictions for skyreels-v2 pipes.
* docs: Remove SkyReels-V2 FLF2V model link and add contributor attribution.
* LTX Video 0.9.8 long multi prompt
* Further align comfyui
- Added the “LTXEulerAncestralRFScheduler” scheduler, aligned with [sample_euler_ancestral_RF](7d6103325e/comfy/k_diffusion/sampling.py (L234))
- Updated the LTXI2VLongMultiPromptPipeline.from_pretrained() method:
- Now uses LTXEulerAncestralRFScheduler by default, for better compatibility with the ComfyUI LTXV workflow.
- Changed the default value of cond_strength from 1.0 to 0.5, aligning with ComfyUI’s default.
- Optimized cross-window overlap blending: moved the latent-space guidance injection to before the UNet and after each step, aligned with[KSamplerX0Inpaint]([ComfyUI/comfy/samplers.py at master · comfyanonymous/ComfyUI](https://github.com/comfyanonymous/ComfyUI/blob/master/comfy/samplers.py#L391))
- Adjusted the default value of skip_steps_sigma_threshold to 1.
* align with diffusers contribute rule
* Add new pipelines and update imports
* Enhance LTXI2VLongMultiPromptPipeline with noise rescaling
Refactor LTXI2VLongMultiPromptPipeline to improve documentation and add noise rescaling functionality.
* Clean up comments in scheduling_ltx_euler_ancestral_rf.py
Removed design notes and limitations from the implementation.
* Enhance video generation example with scheduler
Updated LTXI2VLongMultiPromptPipeline example to include LTXEulerAncestralRFScheduler for ComfyUI parity.
* clean up
* style
* copies
* import ltx scheduler
* copies
* fix
* fix more
* up up
* up up up
* up upup
* Apply suggestions from code review
* Update docs/source/en/api/pipelines/ltx_video.md
* Update docs/source/en/api/pipelines/ltx_video.md
---------
Co-authored-by: yiyixuxu <yixu310@gmail.com>
* [Flux.1] improve pos embed for ascend npu by setting it back to npu computation.
* [Flux.2] improve pos embed for ascend npu by setting it back to npu computation.
* [LongCat-Image] improve pos embed for ascend npu by setting it back to npu computation.
* [Ovis-Image] improve pos embed for ascend npu by setting it back to npu computation.
* Remove unused import of is_torch_npu_available
---------
Co-authored-by: zhangtao <zhangtao529@huawei.com>
