…follow up on huggingface#11873 (huggingface#12264)

* propagate fixes from huggingface#11873 to flux script

* propagate fixes from huggingface#11873 to flux script

* propagate fixes from huggingface#11873 to flux script

* Apply style fixes

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

…ce#12266)

update

…gingface#12236)

* feat: try loading fa3 using kernels when available.

* up

* change to Hub.

* up

* up

* up

* switch env var.

* up

* up

* up

* up

* up

* up

* refresh

* feedback

* feedback

* supported models

* fix

* initial commit

* update

* updates

* update

* update

* update

* update

* update

* update

* addressed PR comments

* update

* addressed PR comments

* update

* update

* update

* update

* update

* update

* updates

* update

* update

* addressed PR comments

* updates

* code formatting

* update

* addressed PR comments

* addressed PR comments

* addressed PR comments

* addressed PR comments

* fix docs and dependencies

* fixed dependency test

---------

Co-authored-by: Sayak Paul <[email protected]>

* feat: add a test for aot.

* up

* Add AttentionMixin to WanVACETransformer3DModel

to enable methods like `set_attn_processor()`.

* Import AttentionMixin in transformer_wan_vace.py

Special thanks to @tolgacangoz 🙇‍♂️

Signed-off-by: co63oc <[email protected]>

init

init

Co-authored-by: Sayak Paul <[email protected]>

* init

* fix

* feedback

* feedback

* add qwen modular

* add qwen-image-cn-inpaint


---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: yiyixuxu <[email protected]>

Co-authored-by: J石页 <[email protected]>
Co-authored-by: Sayak Paul <[email protected]>

* Update utils.py

not perfect but works

engine:
https://github.com/calcuis/gguf-connector/blob/main/src/gguf_connector/quant2c.py


inference example(s):
https://github.com/calcuis/gguf-connector/blob/main/src/gguf_connector/k6.py
https://github.com/calcuis/gguf-connector/blob/main/src/gguf_connector/k5.py

gguf file sample(s):
https://huggingface.co/calcuis/kontext-gguf/tree/main
https://huggingface.co/calcuis/krea-gguf/tree/main

* Apply style fixes

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

…2290)

adjust criteria for XPU

Signed-off-by: Liu, Kaixuan <[email protected]>
Co-authored-by: Aryan <[email protected]>

…ace#12283)

* feat: support group offloading at the pipeline level.

* add tests

* up

* [docs] Pipeline group offloading (huggingface#12286)

init

Co-authored-by: Sayak Paul <[email protected]>

---------

Co-authored-by: Steven Liu <[email protected]>

fix flux modular pipelines for t2i and i2i

)

* add

* add a test

…ingface#12309)

fix the device for textencoder

…mponents (huggingface#12234)

* allow non list components_to_quantize.

* up

* Apply suggestions from code review

* Apply suggestions from code review

Co-authored-by: Steven Liu <[email protected]>

* [docs] components_to_quantize (huggingface#12287)

init

Co-authored-by: Sayak Paul <[email protected]>

---------

Co-authored-by: Steven Liu <[email protected]>

Co-authored-by: YiYi Xu <[email protected]>

…gface#12271)

* deprecate slicing from flux pipeline.

* propagate.

* tiling

* up

* up

* Use SDP on BF16 in GPU/HPU migration

Signed-off-by: Daniel Socek <[email protected]>

* Formatting fix for enabling SDP with BF16 precision on HPU

Signed-off-by: Daniel Socek <[email protected]>

---------

Signed-off-by: Daniel Socek <[email protected]>

* support Wan2.2-VACE-Fun-A14B

* support Wan2.2-VACE-Fun-A14B

* support Wan2.2-VACE-Fun-A14B

* Apply style fixes

* test

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

* fixed bug in defining embed dim

* matched 1d temb process to 2d

* Update src/diffusers/models/unets/unet_1d.py

Co-authored-by: Dhruv Nair <[email protected]>

---------

Co-authored-by: Dhruv Nair <[email protected]>

* Added LucyEditPipeline

* add import & stype

missing copied from

* Fix example doc string

---------

Co-authored-by: yiyixuxu <[email protected]>

* Update autoencoder_kl_wan.py

When using the Wan2.2 VAE, the spatial compression ratio calculated here is incorrect. It should be 16 instead of 8. Pass it in directly via the config to ensure it’s correct here.

* Update autoencoder_kl_wan.py

* fix hidream type hint

* fix hunyuan-video type hint

* fix many type hint

* fix many type hint errors

* fix many type hint errors

* fix many type hint errors

* make stype & make quality

…g with empty dim. (huggingface#12770)

* Refactor image padding logic to pervent zero tensor in transformer_z_image.py

* Apply style fixes

* Add more support to fix repeat bug on tpu devices.

* Fix for dynamo compile error for multi if-branches.

---------

Co-authored-by: Mingjia Li <[email protected]>
Co-authored-by: Mingjia Li <[email protected]>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

…orking properly (huggingface#12721)

* Fixes huggingface#12673.

    Wrong default_stream is used. leading to wrong execution order when record_steram is enabled.

* update

* Update test

---------

Co-authored-by: Sayak Paul <[email protected]>

…2765)

* start varlen variants for attn backend kernels.

* maybe unflatten heads.

* updates

* remove unused function.

* doc

* up

* remove attn_processors property

* more

* up

* up more.

* up

* add AttentionMixin to AuraFlow.

* up

* up

* up

* up

* update

* update

* Revert "update"

This reverts commit 7390638.

* Revert "update"

This reverts commit 21a03f9.

* update

* update

* update

* update

* update

* add transformer pipeline first version


---------

Co-authored-by: Álvaro Somoza <[email protected]>
Co-authored-by: YiYi Xu <[email protected]>
Co-authored-by: Charles <[email protected]>
Co-authored-by: Sayak Paul <[email protected]>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: dmitrienkoae <[email protected]>
Co-authored-by: nvvaulin <[email protected]>

…12639)

* Fix(peft): Re-apply group offloading after deleting adapters

* Test: Add regression test for group offloading + delete_adapters

* Test: Add assertions to verify output changes after deletion

* Test: Add try/finally to clean up group offloading hooks

---------

Co-authored-by: Sayak Paul <[email protected]>

fix hunuyanvideo 1.5 offloading tests.

…gingface#12741)

* start zimage model tests.

* up

* up

* up

* up

* up

* up

* up

* up

* up

* up

* up

* up

* Revert "up"

This reverts commit bca3e27.

* expand upon compilation failure reason.

* Update tests/models/transformers/test_models_transformer_z_image.py

Co-authored-by: dg845 <[email protected]>

* reinitialize the padding tokens to ones to prevent NaN problems.

* updates

* up

* skipping ZImage DiT tests

* up

* up

---------

Co-authored-by: dg845 <[email protected]>

* Z-Image-Turbo `from_single_file`

* compute_dtype

* -device cast

…uggingface#12767)

refactor: add type hints and update docstrings for UniPCMultistepScheduler parameters and methods.

…encode (huggingface#12753)

fix spatial compression ratio compute error for AutoEncoderKLWan

Co-authored-by: lirui.926 <[email protected]>

up

Co-authored-by: Álvaro Somoza <[email protected]>

…mentation (huggingface#12791)

fix timestepembeddings downscale_freq_shift to be consitant with Photoroom's original code

…ne layers (huggingface#12692)

* fix: group offloading to support standalone computational layers in block-level offloading

* test: for models with standalone and deeply nested layers in block-level offloading

* feat: support for block-level offloading in group offloading config

* fix: group offload block modules to AutoencoderKL and AutoencoderKLWan

* fix: update group offloading tests to use AutoencoderKL and adjust input dimensions

* refactor: streamline block offloading logic

* Apply style fixes

* update tests

* update

* fix for failing tests

* clean up

* revert to use skip_keys

* clean up

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Dhruv Nair <[email protected]>

* initial

* toctree

* fix

* apply review and fix

* Update docs/source/en/api/pipelines/z_image.md

Co-authored-by: Steven Liu <[email protected]>

* Update docs/source/en/api/pipelines/z_image.md

Co-authored-by: Steven Liu <[email protected]>

* Update docs/source/en/api/pipelines/z_image.md

Co-authored-by: Steven Liu <[email protected]>

---------

Co-authored-by: Steven Liu <[email protected]>

up

…ggingface#12796)

* feat: Add `flow_prediction` to `prediction_type`, introduce `use_flow_sigmas`, `flow_shift`, `use_dynamic_shifting`, and `time_shift_type` parameters, and refine type hints for various arguments.

* style: reformat argument wrapping in `_convert_to_beta` and `index_for_timestep` method signatures.

* init taylor_seer cache

* make compatible with any tuple size returned

* use logger for printing, add warmup feature

* still update in warmup steps

* refractor, add docs

* add configurable cache, skip compute module

* allow special cache ids only

* add stop_predicts (cooldown)

* update docs

* apply ruff

* update to handle multple calls per timestep

* refractor to use state manager

* fix format & doc

* chores: naming, remove redundancy

* add docs

* quality & style

* fix taylor precision

* Apply style fixes

* add tests

* Apply style fixes

* Remove TaylorSeerCacheTesterMixin from flux2 tests

* rename identifiers, use more expressive taylor predict loop

* torch compile compatible

* Apply style fixes

* Update src/diffusers/hooks/taylorseer_cache.py

Co-authored-by: Dhruv Nair <[email protected]>

* update docs

* make fix-copies

* fix example usage.

* remove tests on flux kontext

---------

Co-authored-by: toilaluan <[email protected]>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Dhruv Nair <[email protected]>
Co-authored-by: Sayak Paul <[email protected]>

Update the naming

Co-authored-by: Sayak Paul <[email protected]>

* add post init for safty checker

Signed-off-by: jiqing-feng <[email protected]>

* check transformers version before post init

Signed-off-by: jiqing-feng <[email protected]>

* Apply style fixes

---------

Signed-off-by: jiqing-feng <[email protected]>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

* support step-distilled

* style

* Add ZImageImg2ImgPipeline

Updated the pipeline structure to include ZImageImg2ImgPipeline
    alongside ZImagePipeline.
Implemented the ZImageImg2ImgPipeline class for image-to-image
    transformations, including necessary methods for
    encoding prompts, preparing latents, and denoising.
Enhanced the auto_pipeline to map the new ZImageImg2ImgPipeline
    for image generation tasks.
Added unit tests for ZImageImg2ImgPipeline to ensure
    functionality and performance.
Updated dummy objects to include ZImageImg2ImgPipeline for
    testing purposes.

* Address review comments for ZImageImg2ImgPipeline

- Add `# Copied from` annotations to encode_prompt and _encode_prompt
- Add ZImagePipeline to auto_pipeline.py for AutoPipeline support

* Add ZImage pipeline documentation

---------

Co-authored-by: YiYi Xu <[email protected]>
Co-authored-by: Álvaro Somoza <[email protected]>

* Reimplement img2seq & seq2img in PRX to enable ONNX build without Col2Im (incompatible with TensorRT).

* Apply style fixes

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Sayak Paul <[email protected]>

…py (huggingface#12798)

feat: add flow sigmas, dynamic shifting, and refine type hints in DPMSolverSinglestepScheduler