GitLab

[Sana] Add Sana, including `SanaPipeline`, `SanaPAGPipeline`, `LinearAttentionProcessor`, `Flow-based DPM-sovler` and so on. (#9982)

* first add a script for DC-AE;

* DC-AE init

* replace triton with custom implementation

* 1. rename file and remove un-used codes;

* no longer rely on omegaconf and dataclass

* replace custom activation with diffuers activation

* remove dc_ae attention in attention_processor.py

* iinherit from ModelMixin

* inherit from ConfigMixin

* dc-ae reduce to one file

* update downsample and upsample

* clean code

* support DecoderOutput

* remove get_same_padding and val2tuple

* remove autocast and some assert

* update ResBlock

* remove contents within super().__init__

* Update src/diffusers/models/autoencoders/dc_ae.py
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* remove opsequential

* update other blocks to support the removal of build_norm

* remove build encoder/decoder project in/out

* remove inheritance of RMSNorm2d from LayerNorm

* remove reset_parameters for RMSNorm2d
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* remove device and dtype in RMSNorm2d __init__
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/autoencoders/dc_ae.py
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/autoencoders/dc_ae.py
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Update src/diffusers/models/autoencoders/dc_ae.py
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* remove op_list & build_block

* remove build_stage_main

* change file name to autoencoder_dc

* move LiteMLA to attention.py

* align with other vae decode output;

* add DC-AE into init files;

* update

* make quality && make style;

* quick push before dgx disappears again

* update

* make style

* update

* update

* fix

* refactor

* refactor

* refactor

* update

* possibly change to nn.Linear

* refactor

* make fix-copies

* replace vae with ae

* replace get_block_from_block_type to get_block

* replace downsample_block_type from Conv to conv for consistency

* add scaling factors

* incorporate changes for all checkpoints

* make style

* move mla to attention processor file; split qkv conv to linears

* refactor

* add tests

* from original file loader

* add docs

* add standard autoencoder methods

* combine attention processor

* fix tests

* update

* minor fix

* minor fix

* minor fix & in/out shortcut rename

* minor fix

* make style

* fix paper link

* update docs

* update single file loading

* make style

* remove single file loading support; todo for DN6

* Apply suggestions from code review
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* add abstract

* 1. add DCAE into diffusers;
2. make style and make quality;

* add DCAE_HF into diffusers;

* bug fixed;

* add SanaPipeline, SanaTransformer2D into diffusers;

* add sanaLinearAttnProcessor2_0;

* first update for SanaTransformer;

* first update for SanaPipeline;

* first success run SanaPipeline;

* model output finally match with original model with the same intput;

* code update;

* code update;

* add a flow dpm-solver scripts

* 🎉[important update]
1. Integrate flow-dpm-sovler into diffusers;
2. finally run successfully on both `FlowMatchEulerDiscreteScheduler` and `FlowDPMSolverMultistepScheduler`;

* 🎉🔧

[important update & fix huge bugs!!]
1. add SanaPAGPipeline & several related Sana linear attention operators;
2. `SanaTransformer2DModel` not supports multi-resolution input;
2. fix the multi-scale HW bugs in SanaPipeline and SanaPAGPipeline;
3. fix the flow-dpm-solver set_timestep() init `model_output` and `lower_order_nums` bugs;

* remove prints;

* add convert sana official checkpoint to diffusers format Safetensor.

* Update src/diffusers/models/transformers/sana_transformer_2d.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/models/transformers/sana_transformer_2d.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/models/transformers/sana_transformer_2d.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/pipelines/pag/pipeline_pag_sana.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/models/transformers/sana_transformer_2d.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/models/transformers/sana_transformer_2d.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/pipelines/sana/pipeline_sana.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/pipelines/sana/pipeline_sana.py
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* update Sana for DC-AE's recent commit;

* make style && make quality

* Add StableDiffusion3PAGImg2Img Pipeline + Fix SD3 Unconditional PAG (#9932)

* fix progress bar updates in SD 1.5 PAG Img2Img pipeline

---------
Co-authored-by: Vinh H. Pham <phamvinh257@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* make the vae can be None in `__init__` of `SanaPipeline`

* Update src/diffusers/models/transformers/sana_transformer_2d.py
Co-authored-by: hlky <hlky@hlky.ac>

* change the ae related code due to the latest update of DCAE branch;

* change the ae related code due to the latest update of DCAE branch;

* 1. change code based on AutoencoderDC;
2. fix the bug of new GLUMBConv;
3. run success;

* update for solving conversation.

* 1. fix bugs and run convert script success;
2. Downloading ckpt from hub automatically;

* make style && make quality;

* 1. remove un-unsed parameters in init;
2. code update;

* remove test file

* refactor; add docs; add tests; update conversion script

* make style

* make fix-copies

* refactor

* udpate pipelines

* pag tests and refactor

* remove sana pag conversion script

* handle weight casting in conversion script

* update conversion script

* add a processor

* 1. add bf16 pth file path;
2. add complex human instruct in pipeline;

* fix fast \tests

* change gemma-2-2b-it ckpt to a non-gated repo;

* fix the pth path bug in conversion script;

* change grad ckpt to original; make style

* fix the complex_human_instruct bug and typo;

* remove dpmsolver flow scheduler

* apply review suggestions

* change the `FlowMatchEulerDiscreteScheduler` to default `DPMSolverMultistepScheduler` with flow matching scheduler.

* fix the tokenizer.padding_side='right' bug;

* update docs

* make fix-copies

* fix imports

* fix docs

* add integration test

* update docs

* update examples

* fix convert_model_output in schedulers

* fix failing tests

---------
Co-authored-by: Junyu Chen <chenjydl2003@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: chenjy2003 <70215701+chenjy2003@users.noreply.github.com>
Co-authored-by: Aryan <aryan@huggingface.co>
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
Co-authored-by: hlky <hlky@hlky.ac>

* using sd inpaint pipeline and sdxl pag inpaint pipeline to add changes

* using sd inpaint pipeline and sdxl pag inpaint pipeline to add changes

* finished the call function

* added auto pipeline

* merging diffusers

* ready to test

* ready to test

* added copied from and removed unnecessary tests

* make style changes

* doc changes

* updating example doc string

* style fix

* init

* adding imports

* quality

* Update src/diffusers/pipelines/pag/pipeline_pag_sd_inpaint.py

* make

* Update tests/pipelines/pag/test_pag_sd_inpaint.py

* slice and size

* slice

---------
Co-authored-by: Darshil Jariwala <darshiljariwala@Darshils-MacBook-Air.local>
Co-authored-by: Darshil Jariwala <jariwala.darshil2002@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: hlky <hlky@hlky.ac>

* fix progress bar updates in SD 1.5 PAG Img2Img pipeline



---------
Co-authored-by: Vinh H. Pham <phamvinh257@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* added pag to sd img2img pipeline


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

* Add pag to controlnet inpainting pipeline


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

* Added pad controlnet sdxl img2img pipeline

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

* txt2img pag added

* autopipe added, fixed case

* style

* apply suggestions

* added fast tests, added todo tests

* revert dummy objects for kolors

* fix pag dummies

* fix test imports

* update pag tests

* add kolor pag to docs

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

add pag sd3


---------
Co-authored-by: HyoungwonCho <jhw9811@korea.ac.kr>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: crepejung00 <jaewoojung00@naver.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: Aryan <contact.aryanvs@gmail.com>
Co-authored-by: Aryan <aryan@huggingface.co>

* copy hunyuandit pipeline

* pag variant of hunyuan dit

* add tests

* update docs

* make style

* make fix-copies

* Update src/diffusers/pipelines/pag/pag_utils.py

* remove incorrect copied from

* remove pag hunyuan attn procs to resolve conflicts

* add pag attn procs again

* new implementation for pag_utils

* revert pag changes

* add pag refactor back; update pixart sigma

* update pixart pag tests

* apply suggestions from review

Co-Authored-By: yixu310@gmail.com

* make style

* update docs, fix tests

* fix tests

* fix test_components_function since list not accepted as valid __init__ param

* apply patch to fix broken tests
Co-Authored-By: Sayak Paul <spsayakpaul@gmail.com>

* make style

* fix hunyuan tests

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

* feat: add pixart sigma pag.

* inits.

* fixes

* fix

* remove print.

* copy paste methods to the pixart pag mixin

* fix-copies

* add documentation.

* add tests.

* remove correction file.

* remove pag_applied_layers

* empty

* add animatediff pag pipeline

* remove unnecessary print

* make fix-copies

* fix ip-adapter bug

* update docs

* add fast tests and fix bugs

* update

* update

* address review comments

* update ip adapter single test expected slice

* implement test_from_pipe_consistent_config; fix expected slice values

* LoraLoaderMixin->StableDiffusionLoraLoaderMixin; add latest freeinit test

* add pag support sd15 controlnet

* fix quality import

* remove unecessary import

* remove if state

* fix tests

* remove useless function

* add sd1.5 controlnet pag docs

---------
Co-authored-by: anhnct8 <anhnct8@fpt.com>

* add pag to sd pipelines

* first draft


---------
Co-authored-by: yiyixuxu <yixu310@gmail,com>
Co-authored-by: Junhwa Song <ethan9867@gmail.com>
Co-authored-by: Ahn Donghoon (안동훈 / suno) <suno.vivid@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* sd3 controlnet



---------
Co-authored-by: haofanwang <haofanwang.ai@gmail.com>

* Hunyuan Team: add HunyuanDiT related updates


---------
Co-authored-by: XCLiu <liuxc1996@gmail.com>
Co-authored-by: yiyixuxu <yixu310@gmail.com>

---------
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* fix: import bug

* fix

* fix

* fix import utils for lcm

* fix: pixart alpha init

* Fix

---------
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* init pixart alpha pipeline

* fix: import

* script

* script

* script

* add: vae to the pipeline

* add: vae_scale_factor

* add: checkpoint_path

* clean conversion script a bit.

* size embeddings.

* fix: size embedding

* update scrip

* support for interpolation of position embedding.

* support for conditioning.

* ..

* ..

* ..

* final layer

* final layer

* align if encode_prompt

* support for caption embedding

* refactor

* refactor

* refactor

* start cross attention

* start cross attention

* cross_attention_dim

* cross

* cross

* support for resolution and aspect_ratio

* support for caption projection

* refactor patch embeddings

* batch_size

* up

* commit

* commit

* commit.

* squeeze

* squeeze

* squeeze

* squeeze

* squeeze

* squeeze

* squeeze

* squeeze

* squeeze

* squeeze

* squeeze

* squeeze.

* squeeze.

* fix final block./

* fix final block./

* fix final block./

* clean

* fix: interpolation scale.

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging'

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* make --checkpoint_path non-required.

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* remove num_tokens

* timesteps -> timestep

* timesteps -> timestep

* timesteps -> timestep

* timesteps -> timestep

* timesteps -> timestep

* timesteps -> timestep

* debug

* debug

* update conversion script.

* update conversion script.

* update conversion script.

* debug

* debug

* debug

* clean

* debug

* debug

* debug

* debug

* debug

* debug

* debug

* debug

* deug

* debug

* debug

* debug

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* clean

* fix

* fix

* boom

* boom

* some changes

* boom

* save

* up

* remove i

* fix more tests

* DPMSolverMultistepScheduler

* fix

* offloading

* fix conversion script

* fix conversion script

* remove print

* remove support for negative prompt embeds.

* typo.

* remove extra kwargs

* bring conversion script to where it was

* fix

* trying mu luck

* trying my luck again

* again

* again

* again

* clean up

* up

* up

* update example

* support for 512

* remove spacing

* finalize docs.

* test debug

* fix: assertion values.

* debug

* debug

* debug

* fix: repeat

* remove prints.

* Apply suggestions from code review

* Apply suggestions from code review

* Correct more

* Apply suggestions from code review

* Change all

* Clean more

* fix more

* Fix more

* Fix more

* Correct more

* address patrick's comments.

* remove unneeded args

* clean up pipeline.

* sty;e

* make the use of additional conditions better conditioned.

* None better

* dtype

* height and width validation

* add a note about size brackets.

* fix

* spit out slow test outputs.

* fix?

* fix optional test

* fix more

* remove unneeded comment

* debug

---------
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>