Skip to content
Open

13 #1

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
857 commits
Select commit Hold shift + click to select a range
256e010
Improve docstrings and type hints in scheduling_deis_multistep.py (#1…
delmalih Dec 5, 2025
6290fdf
[Feat] TaylorSeer Cache (#12648)
toilaluan Dec 6, 2025
5a74319
Update the TensorRT-ModelOPT to Nvidia-ModelOPT (#12793)
jingyu-ml Dec 8, 2025
f67639b
add post init for safty checker (#12794)
jiqing-feng Dec 8, 2025
671149e
[HunyuanVideo1.5] support step-distilled (#12802)
yiyixuxu Dec 8, 2025
2246d2c
Add ZImageImg2ImgPipeline (#12751)
CalamitousFelicitousness Dec 8, 2025
3d02cd5
[PRX] Improve model compilation (#12787)
WaterKnight1998 Dec 8, 2025
54fa074
Improve docstrings and type hints in scheduling_dpmsolver_singlestep.…
delmalih Dec 8, 2025
07ea078
[Modular]z-image (#12808)
yiyixuxu Dec 9, 2025
8b4722d
Fix Qwen Edit Plus modular for multi-image input (#12601)
sayakpaul Dec 9, 2025
be3c2a0
[WIP] Add Flux2 modular (#12763)
DN6 Dec 10, 2025
6708f5c
[docs] improve distributed inference cp docs. (#12810)
sayakpaul Dec 10, 2025
10e820a
post release 0.36.0 (#12804)
sayakpaul Dec 11, 2025
0eac64c
Update distributed_inference.md to correct syntax (#12827)
sayakpaul Dec 11, 2025
1567243
[lora] Remove lora docs unneeded and add " # Copied from ..." (#12824)
sayakpaul Dec 12, 2025
17c0e79
support CP in native flash attention (#12829)
sywangyi Dec 12, 2025
b8a4cba
[qwen-image] edit 2511 support (#12839)
naykun Dec 15, 2025
0c1ccc0
fix pytest tests/pipelines/pixart_sigma/test_pixart.py::PixArtSigmaPi…
sywangyi Dec 15, 2025
5851928
Support for control-lora (#10686)
lavinal712 Dec 15, 2025
a748a83
Add support for LongCat-Image (#12828)
junqiangwu Dec 15, 2025
5e48f46
fix the prefix_token_len bug (#12845)
junqiangwu Dec 16, 2025
87f7d11
extend TorchAoTest::test_model_memory_usage to other platform (#12768)
sywangyi Dec 17, 2025
f9c1e61
Qwen Image Layered Support (#12853)
naykun Dec 17, 2025
55463f7
Z-Image-Turbo ControlNet (#12792)
hlky Dec 17, 2025
b530968
Cosmos Predict2.5 Base: inference pipeline, scheduler & chkpt convers…
miguelmartin75 Dec 19, 2025
f7753b1
more update in modular (#12560)
yiyixuxu Dec 19, 2025
262ce19
Feature: Add Mambo-G Guidance as Guider (#12862)
MatrixTeam-AI Dec 19, 2025
0c4f6c9
Add `OvisImagePipeline` in `AUTO_TEXT2IMAGE_PIPELINES_MAPPING` (#12876)
alvarobartt Dec 22, 2025
973a077
Cosmos Predict2.5 14b Conversion (#12863)
miguelmartin75 Dec 22, 2025
52766e6
Use `T5Tokenizer` instead of `MT5Tokenizer` (removed in Transformers …
alvarobartt Dec 23, 2025
f6b6a71
Add z-image-omni-base implementation (#12857)
RuoyiDu Dec 24, 2025
1cdb872
fix torchao quantizer for new torchao versions (#12901)
vkuzo Dec 30, 2025
208cda8
fix Qwen Image Transformer single file loading mapping function to be…
mbalabanski Jan 2, 2026
4737806
Z-Image-Turbo from_single_file fix (#12888)
hlky Jan 2, 2026
d0ae34d
chore: fix dev version in setup.py (#12904)
DefTruth Jan 5, 2026
5ffb658
Community Pipeline: Add z-image differential img2img (#12882)
r4inm4ker Jan 5, 2026
0da1aa9
Fix typo in src/diffusers/pipelines/cosmos/pipeline_cosmos2_5_predict…
miguelmartin75 Jan 6, 2026
3138e37
Fix wan 2.1 i2v context parallel (#12909)
DefTruth Jan 6, 2026
7c6d314
fix the use of device_map in CP docs (#12902)
sayakpaul Jan 6, 2026
b6098ca
[core] remove unneeded autoencoder methods when subclassing from `Aut…
sayakpaul Jan 6, 2026
88ffb00
Detect 2.0 vs 2.1 ZImageControlNetModel (#12861)
hlky Jan 6, 2026
db37140
Refactor environment variable assignments in workflow (#12916)
paulinebm Jan 6, 2026
e46354d
Add codeQL workflow (#12917)
paulinebm Jan 6, 2026
417f6b2
Delete .github/workflows/codeql.yml
paulinebm Jan 6, 2026
9b5a244
CodeQL workflow for security analysis
paulinebm Jan 6, 2026
41a6e86
Check for attention mask in backends that don't support it (#12892)
dxqb Jan 6, 2026
ade1059
[Flux.1] improve pos embed for ascend npu by computing on npu (#12897)
zhangtao0408 Jan 6, 2026
98479a9
LTX Video 0.9.8 long multi prompt (#12614)
yaoqih Jan 7, 2026
b4be29b
Add FSDP option for Flux2 (#12860)
leisuzz Jan 7, 2026
8f30bff
Add transformer cache context for SkyReels-V2 pipelines & Update docs…
tolgacangoz Jan 7, 2026
961b9b2
[docs] fix torchao typo. (#12883)
sayakpaul Jan 7, 2026
6fb4c99
Update wan.md to remove unneeded hfoptions (#12890)
sayakpaul Jan 7, 2026
9fb6b89
Improve docstrings and type hints in scheduling_edm_euler.py (#12871)
delmalih Jan 7, 2026
dab000e
[Modular] Video for Mellon (#12924)
asomoza Jan 7, 2026
c10bdd9
Add LTX 2.0 Video Pipelines (#12915)
dg845 Jan 8, 2026
8600b4c
Add environment variables to checkout step (#12927)
paulinebm Jan 8, 2026
b1f06b7
Improve docstrings and type hints in scheduling_consistency_decoder.p…
delmalih Jan 8, 2026
8b9f817
Fix: Remove hardcoded CUDA autocast in Kandinsky 5 to fix import warn…
adi776borate Jan 8, 2026
a812c87
Upgrade GitHub Actions for Node 24 compatibility (#12865)
salmanmkc Jan 9, 2026
91e5134
fix the warning torch_dtype is deprecated (#12841)
msdsm Jan 9, 2026
be38f41
[NPU] npu attention enable ulysses (#12919)
TmacAaron Jan 9, 2026
2f66edc
Torchao floatx version guard (#12923)
howardzhang-cv Jan 9, 2026
08f8b7a
Bugfix for dreambooth flux2 img2img2 (#12825)
leisuzz Jan 9, 2026
1903383
[Modular] qwen refactor (#12872)
yiyixuxu Jan 9, 2026
3981c95
[modular] Tests for custom blocks in modular diffusers (#12557)
sayakpaul Jan 9, 2026
d568c97
[chore] remove controlnet implementations outside controlnet module. …
sayakpaul Jan 9, 2026
441b69e
[core] Handle progress bar and logging in distributed environments (#…
sayakpaul Jan 9, 2026
d36564f
Improve docstrings and type hints in scheduling_consistency_models.py…
delmalih Jan 9, 2026
632765a
[Feature] MultiControlNet support for SD3Impainting (#11251)
ishan-modi Jan 9, 2026
6441694
Laplace Scheduler for DDPM (#11320)
gapatron Jan 9, 2026
57e57cf
Store vae.config.scaling_factor to prevent missing attr reference (sd…
Teriks Jan 9, 2026
2dc9d2a
Add thread-safe wrappers for components in pipeline (examples/server-…
FredyRivera-dev Jan 9, 2026
2bb640f
[Research] Latent Perceptual Loss (LPL) for Stable Diffusion XL (#11573)
kashif Jan 9, 2026
4b64b56
Change timestep device to cpu for xla (#11501)
bhavya01 Jan 9, 2026
b7a8158
[LoRA] add lora_alpha to sana README (#11780)
linoytsaban Jan 9, 2026
4fb44bd
Fix wrong param types, docs, and handles noise=None in scale_noise of…
Promisery Jan 9, 2026
5794fff
[docs] Remote inference (#12372)
stevhliu Jan 9, 2026
df6516a
Align HunyuanVideoConditionEmbedding with CombinedTimestepGuidanceTex…
samutamm Jan 9, 2026
1791306
[Fix] syntax in QwenImageEditPlusPipeline (#12371)
SahilCarterr Jan 9, 2026
2a6cdc0
Fix ftfy name error in Wan pipeline (#12314)
dsocek Jan 10, 2026
a3cc0e7
[modular] error early in `enable_auto_cpu_offload` (#12578)
sayakpaul Jan 10, 2026
02c7adc
[ChronoEdit] support multiple loras (#12679)
zhangjiewu Jan 10, 2026
d44b5f8
fix how `is_fsdp` is determined (#12960)
sayakpaul Jan 10, 2026
ed6e5ec
[LoRA] add LoRA support to LTX-2 (#12933)
sayakpaul Jan 10, 2026
2120c30
Fix: typo in autoencoder_dc.py (#12687)
tvelovraf Jan 10, 2026
418313b
[Modular] better docstring (#12932)
yiyixuxu Jan 10, 2026
4ac2b4a
[docs] polish caching docs. (#12684)
sayakpaul Jan 10, 2026
5b20211
Fix typos (#12705)
omahs Jan 10, 2026
b86bd99
Fix link to diffedit implementation reference (#12708)
JuanFKurucz Jan 10, 2026
dad5cb5
Fix QwenImage txt_seq_lens handling (#12702)
kashif Jan 12, 2026
29a930a
Bugfix for flux2 img2img2 prediction (#12855)
leisuzz Jan 12, 2026
f1a93c7
Add Flag to `PeftLoraLoaderMixinTests` to Enable/Disable Text Encoder…
dg845 Jan 13, 2026
9d68742
Add Unified Sequence Parallel attention (#12693)
Bissmella Jan 13, 2026
3114f6a
[Modular] Changes for using WAN I2V (#12959)
asomoza Jan 13, 2026
7299121
Z rz rz rz rz rz rz r cogview (#12973)
sayakpaul Jan 13, 2026
3c70440
Update distributed_inference.md to reposition sections (#12971)
sayakpaul Jan 13, 2026
7feb4fc
[chore] make transformers version check stricter for glm image. (#12974)
sayakpaul Jan 14, 2026
d7fa445
Remove 8bit device restriction (#12972)
SunMarc Jan 14, 2026
1ecfbfe
`disable_mmap` in pipeline `from_pretrained` (#12854)
hlky Jan 14, 2026
d8f4dd2
[Modular] mellon utils (#12978)
yiyixuxu Jan 15, 2026
b351be2
LongCat Image pipeline: Allow offloading/quantization of text_encoder…
Yahweasel Jan 15, 2026
5efb81f
Add `ChromaInpaintPipeline` (#12848)
hameerabbasi Jan 15, 2026
7f43cb1
fix Qwen-Image series context parallel (#12970)
DefTruth Jan 15, 2026
61f1756
Flux2 klein (#12982)
yiyixuxu Jan 15, 2026
f112eab
[modular] fix a bug in mellon param & improve docstrings (#12980)
yiyixuxu Jan 15, 2026
74654df
add klein docs. (#12984)
sayakpaul Jan 16, 2026
8af8e86
LTX 2 Single File Support (#12983)
dg845 Jan 16, 2026
ebf891a
[core] gracefully error out when attn-backend x cp combo isn't suppor…
sayakpaul Jan 16, 2026
9fedfe5
Improve docstrings and type hints in scheduling_cosine_dpmsolver_mult…
delmalih Jan 16, 2026
3996788
[Docs] Replace root CONTRIBUTING.md with symlink to source docs (#12986)
delmalih Jan 16, 2026
76f51a5
make style && make quality
sayakpaul Jan 19, 2026
75edff9
Revert "make style && make quality"
sayakpaul Jan 19, 2026
29b15f4
[chore] make style to push new changes. (#12998)
sayakpaul Jan 19, 2026
d7a1c31
Fibo edit pipeline (#12930)
galbria Jan 19, 2026
4b843c8
Fix variable name in docstring for PeftAdapterMixin.set_adapters (#13…
geekuillaume Jan 20, 2026
ec37629
Improve docstrings and type hints in scheduling_ddim_cogvideox.py (#1…
delmalih Jan 20, 2026
7a02fad
[scheduler] Support custom sigmas in UniPCMultistepScheduler (#12109)
a-r-r-o-w Jan 22, 2026
699297f
feat: accelerate longcat-image with regional compile (#13019)
lgyStoic Jan 22, 2026
1d32b19
Improve docstrings and type hints in scheduling_ddim_flax.py (#13010)
delmalih Jan 22, 2026
d4f97d1
Improve docstrings and type hints in scheduling_ddim_inverse.py (#13020)
delmalih Jan 22, 2026
bff672f
fix Dockerfiles for cuda and xformers. (#13022)
sayakpaul Jan 23, 2026
1eb40c6
Resnet only use contiguous in training mode. (#12977)
jiqing-feng Jan 23, 2026
981eb80
feat: add qkv projection fuse for longcat transformers (#13021)
lgyStoic Jan 23, 2026
6e24cd8
Improve docstrings and type hints in scheduling_ddim_parallel.py (#13…
delmalih Jan 23, 2026
e8e88ff
Improve docstrings and type hints in scheduling_ddpm_flax.py (#13024)
delmalih Jan 23, 2026
a7cb14e
Improve docstrings and type hints in scheduling_ddpm_parallel.py (#13…
delmalih Jan 25, 2026
2af7baa
Remove `*pooled_*` mentions from Chroma inpaint (#13026)
hameerabbasi Jan 26, 2026
956bdcc
Flag Flax schedulers as deprecated (#13031)
delmalih Jan 26, 2026
b1c77f6
[modular] add auto_docstring & more doc related refactors (#12958)
yiyixuxu Jan 27, 2026
d9959bd
Upgrade GitHub Actions to latest versions (#12866)
salmanmkc Jan 27, 2026
53279ef
[From Single File] support `from_single_file` method for `WanAnimateT…
samadwar Jan 27, 2026
71a865b
Fix: Cosmos2.5 Video2World frame extraction and add default negative …
adi776borate Jan 27, 2026
22ac6fa
[GLM-Image] Add batch support for GlmImagePipeline (#13007)
JaredforReal Jan 27, 2026
d54669a
[Qwen] avoid creating attention masks when there is no padding (#12987)
kashif Jan 27, 2026
53d8a1e
[modular]support klein (#13002)
yiyixuxu Jan 28, 2026
ef91301
[QwenImage] fix prompt isolation tests (#13042)
sayakpaul Jan 28, 2026
2ac39ba
fast tok update (#13036)
itazap Jan 28, 2026
2c669e8
change to CUDA 12.9. (#13045)
sayakpaul Jan 28, 2026
74a0f0b
remove torchao autoquant from diffusers docs (#13048)
vkuzo Jan 28, 2026
0ab2124
docs: improve docstring scheduling_dpm_cogvideox.py (#13044)
delmalih Jan 28, 2026
a58d0b9
Fix Wan/WanI2V patchification (#13038)
Jayce-Ping Jan 29, 2026
a2ea45a
LTX2 distilled checkpoint support (#12934)
rootonchair Jan 29, 2026
e7de7d8
[wan] fix layerwise upcasting tests on CPU (#13039)
sayakpaul Jan 29, 2026
314cfdd
[ci] uniform run times and wheels for pytorch cuda. (#13047)
sayakpaul Jan 29, 2026
1be2f7e
docs: fix grammar in fp16_safetensors CLI warning (#13040)
Olexandr88 Jan 29, 2026
f5b6b66
[wan] fix wan 2.2 when either of the transformers isn't present. (#13…
sayakpaul Jan 29, 2026
6a1904e
[bug fix] GLM-Image fit new `get_image_features` API (#13052)
JaredforReal Jan 30, 2026
ec6b2bc
Fix aiter availability check (#13059)
lauri9 Jan 30, 2026
769a1f3
[Modular]add a real quick start guide (#13029)
yiyixuxu Jan 31, 2026
973e334
feat: support Ulysses Anything Attention (#12996)
DefTruth Feb 2, 2026
0b76728
Refactor Model Tests (#12822)
DN6 Feb 2, 2026
b712042
[Flux2] Fix LoRA loading for Flux2 Klein by adaptively enumerating tr…
songkey Feb 2, 2026
ebd06f9
[Modular] loader related (#13025)
yiyixuxu Feb 3, 2026
6d4fc6b
[Modular] mellon doc etc (#13051)
yiyixuxu Feb 3, 2026
1b8fc6c
[modular] change the template modular pipeline card (#13072)
sayakpaul Feb 4, 2026
430c557
Add support for Magcache (#12744)
AlanPonnachan Feb 4, 2026
90818e8
[docs] Fix syntax error in quantization configuration (#13076)
sayakpaul Feb 4, 2026
03af690
docs: improve docstring scheduling_dpmsolver_multistep_inverse.py (#1…
delmalih Feb 4, 2026
9fe0a9c
[core] make flux hidden states contiguous (#13068)
sayakpaul Feb 5, 2026
a3dcd98
[core] make qwen hidden states contiguous to make torchao happy. (#13…
sayakpaul Feb 5, 2026
99e2cff
Feature/zimage inpaint pipeline (#13006)
CalamitousFelicitousness Feb 5, 2026
ca79f8c
GGUF fix for unquantized types when using unquantize kernels (#12498)
dxqb Feb 6, 2026
44b8201
docs: improve docstring scheduling_dpmsolver_multistep_inverse.py (#1…
delmalih Feb 6, 2026
10dc589
[modular]simplify components manager doc (#13088)
yiyixuxu Feb 6, 2026
09dca38
ZImageControlNet cfg (#13080)
hlky Feb 7, 2026
fd705bd
[Modular] refactor Wan: modular pipelines by task etc (#13063)
yiyixuxu Feb 7, 2026
44f4dc0
[Modular] guard `ModularPipeline.blocks` attribute (#13014)
yiyixuxu Feb 9, 2026
baaa8d0
LTX 2 Improve `encode_video` by Accepting More Input Types (#13057)
dg845 Feb 9, 2026
8933686
Z image lora training (#13056)
linoytsaban Feb 9, 2026
20efb79
[modular] add modular tests for Z-Image and Wan (#13078)
sayakpaul Feb 9, 2026
bedc67c
[Docs] Add guide for AutoModel with custom code (#13099)
DN6 Feb 10, 2026
5bf248d
[SkyReelsV2] Fix ftfy import (#13113)
asomoza Feb 10, 2026
4d00980
[lora] fix non-diffusers lora key handling for flux2 (#13119)
sayakpaul Feb 11, 2026
c3a4cd1
[CI] Refactor Wan Model Tests (#13082)
DN6 Feb 11, 2026
64e2adf
docs: improve docstring scheduling_edm_dpmsolver_multistep.py (#13122)
delmalih Feb 11, 2026
d324839
[Fix]Allow `prompt` and `prior_token_ids` to be provided simultaneous…
JaredforReal Feb 11, 2026
06a0f98
docs: improve docstring scheduling_flow_match_euler_discrete.py (#13127)
delmalih Feb 12, 2026
a181616
Cosmos Transfer2.5 inference pipeline: general/{seg, depth, blur, edg…
miguelmartin75 Feb 12, 2026
ed77a24
[modular] add tests for robust model loading. (#13120)
sayakpaul Feb 12, 2026
985d83c
Fix LTX-2 Inference when `num_videos_per_prompt > 1` and CFG is Enabl…
dg845 Feb 12, 2026
427472e
[CI] Fix `setuptools` `pkg_resources` Errors (#13129)
dg845 Feb 12, 2026
5f3ea22
docs: improve docstring scheduling_flow_match_heun_discrete.py (#13130)
delmalih Feb 12, 2026
277e305
[CI] Fix `setuptools` `pkg_resources` Bug for PR GPU Tests (#13132)
dg845 Feb 13, 2026
76af013
fix cosmos transformer typing. (#13134)
sayakpaul Feb 13, 2026
2843b3d
Sunset Python 3.8 & get rid of explicit `typing` exports where possib…
sayakpaul Feb 13, 2026
8abcf35
feat: implement apply_lora_scale to remove boilerplate. (#12994)
sayakpaul Feb 13, 2026
3c1c62e
[docs] fix ltx2 i2v docstring. (#13135)
sayakpaul Feb 14, 2026
6141ae2
[Modular] add different pipeine blocks to init (#13145)
yiyixuxu Feb 14, 2026
5b00a18
fix MT5Tokenizer (#13146)
yiyixuxu Feb 14, 2026
19ab0ec
fix guider (#13147)
yiyixuxu Feb 14, 2026
3c7506b
[Modular] update doc for `ModularPipeline` (#13100)
yiyixuxu Feb 14, 2026
c919ec0
[Modular] add explicit workflow support (#13028)
yiyixuxu Feb 15, 2026
b0dc51d
[LTX2] Fix wrong lora mixin (#13144)
asomoza Feb 15, 2026
59e7a46
[Pipelines] Remove k-diffusion (#13152)
DN6 Feb 16, 2026
e390646
[tests] accept recompile_limit from the user in tests (#13150)
sayakpaul Feb 16, 2026
35086ac
[core] support device type device_maps to work with offloading. (#12811)
sayakpaul Feb 16, 2026
bcbbded
[Bug] Fix QwenImageEditPlus Series on NPU (#13017)
zhangtao0408 Feb 17, 2026
f81e653
[CI] Add ftfy as a test dependency (#13155)
DN6 Feb 18, 2026
64734b2
docs: improve docstring scheduling_flow_match_lcm.py (#13160)
delmalih Feb 18, 2026
6875490
[docs] add docs for qwenimagelayered (#13158)
stevhliu Feb 18, 2026
a577ec3
Flux2: Tensor tuples can cause issues for checkpointing (#12777)
dxqb Feb 19, 2026
53e1d0e
[CI] Revert `setuptools` CI Fix as the Failing Pipelines are Deprecat…
dg845 Feb 19, 2026
fe78a7b
Fix `ftfy` import for PRX Pipeline (#13154)
dg845 Feb 19, 2026
99daaa8
[core] Enable CP for kernels-based attention backends (#12812)
sayakpaul Feb 19, 2026
f8d3db9
remove deps related to test from ci (#13164)
sayakpaul Feb 20, 2026
db2d7e7
[CI] Fix new LoRAHotswap tests (#13163)
DN6 Feb 20, 2026
01de02e
[gguf][torch.compile time] Convert to plain tensor earlier in dequant…
anijain2305 Feb 20, 2026
a80b192
Support Flux Klein peft (fal) lora format (#13169)
asomoza Feb 21, 2026
f1e5914
Fix T5GemmaEncoder loading for transformers 5.x composite T5GemmaConf…
DavidBert Feb 23, 2026
4890e9b
Allow Automodel to use `from_config` with custom code. (#13123)
DN6 Feb 23, 2026
7ab2011
Fix AutoModel `typing` Import Error (#13178)
dg845 Feb 24, 2026
5e94d62
migrate to `transformers` v5 (#12976)
sayakpaul Feb 24, 2026
1f6ac1c
fix: graceful fallback when attention backends fail to import (#13060)
sym-bot Feb 24, 2026
aac94be
[docs] Fix torchrun command argument order in docs (#13181)
sayakpaul Feb 24, 2026
3105848
[attention backends] use dedicated wrappers from fa3 for cp. (#13165)
sayakpaul Feb 25, 2026
212db7b
Cosmos Transfer2.5 Auto-Regressive Inference Pipeline (#13114)
miguelmartin75 Feb 26, 2026
97c2c6e
Fix wrong `do_classifier_free_guidance` threshold in ZImagePipeline (…
kirillsst Feb 26, 2026
47455bd
Fix Flash Attention 3 interface for new FA3 return format (#13173)
veeceey Feb 26, 2026
40e9645
Fix LTX-2 image-to-video generation failure in two stages generation …
Songrui625 Feb 27, 2026
5910a1c
Fixing Kohya loras loading: Flux.1-dev loras with TE ("lora_te1_" pre…
christopher5106 Feb 27, 2026
680076f
[Modular] update the auto pipeline blocks doc (#13148)
yiyixuxu Feb 27, 2026
9b97932
[tests] consistency tests for modular index (#13192)
sayakpaul Feb 28, 2026
3918824
[modular] fallback to default_blocks_name when loading base block cla…
yiyixuxu Feb 28, 2026
fcdd759
[chore] updates in the pypi publication workflow. (#12805)
sayakpaul Mar 2, 2026
5b54496
[tests] enable cpu offload test in torchao without compilation. (#12704)
sayakpaul Mar 2, 2026
3d90855
remove db utils from benchmarking (#13199)
sayakpaul Mar 2, 2026
e7fe4ce
[AutoModel] Fix bug with subfolders and local model paths when loadin…
DN6 Mar 2, 2026
3fd14f1
[AutoModel] Allow registering `auto_map` to model config (#13186)
DN6 Mar 2, 2026
84ff061
[Modular] Save Modular Pipeline weights to Hub (#13168)
DN6 Mar 2, 2026
c2fdd2d
docs: improve docstring scheduling_ipndm.py (#13198)
delmalih Mar 2, 2026
47e8faf
Clean up accidental files (#13202)
DN6 Mar 2, 2026
bbbcdd8
[modular]Update model card to include workflow (#13195)
yiyixuxu Mar 3, 2026
1fe688a
[modular] not pass trust_remote_code to external repos (#13204)
yiyixuxu Mar 3, 2026
4a2833c
[Modular] implement requirements validation for custom blocks (#12196)
sayakpaul Mar 4, 2026
8879824
cogvideo example: Distribute VAE video encoding across processes in C…
jiqing-feng Mar 4, 2026
06ccde9
Fix group-offloading bug (#13211)
SHYuanBest Mar 4, 2026
33f785b
Add Helios-14B Video Generation Pipelines (#13208)
dg845 Mar 4, 2026
5570f81
[Z-Image] Fix more `do_classifier_free_guidance` thresholds (#13212)
asomoza Mar 4, 2026
3902145
[lora] fix zimage lora conversion to support for more lora. (#13209)
sayakpaul Mar 5, 2026
20364fe
adding lora support to z-image controlnet pipelines (#13200)
christopher5106 Mar 5, 2026
ab6040a
Add LTX2 Condition Pipeline (#13058)
dg845 Mar 5, 2026
ae5881b
Fix Helios paper link in documentation (#13213)
SHYuanBest Mar 5, 2026
29b9109
[attention backends] change to updated repo and version. (#13161)
sayakpaul Mar 5, 2026
8ec0a5c
feat: implement rae autoencoder. (#13046)
Ando233 Mar 5, 2026
46bd005
Convert tensors to float in Helios's optimized_scale function (#13214)
SHYuanBest Mar 6, 2026
e747fe4
Fix wrapped transformer config access in Flux2 Klein training (#13219)
tcaimm Mar 6, 2026
e1b5db5
post release 0.37.0 (#13215)
sayakpaul Mar 7, 2026
9254417
Fix Helios Context Parallelism (#13223)
SHYuanBest Mar 8, 2026
bd7a7a0
Optimize Helios docs (#13222)
SHYuanBest Mar 8, 2026
7f92d81
Add VidTok AutoEncoders (#11261)
annitang1997 Mar 9, 2026
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
40 changes: 23 additions & 17 deletions .github/workflows/benchmark.yml
Original file line number Diff line number Diff line change
Expand Up @@ -7,55 +7,61 @@ on:

env:
DIFFUSERS_IS_CI: yes
HF_HUB_ENABLE_HF_TRANSFER: 1
HF_XET_HIGH_PERFORMANCE: 1
HF_HOME: /mnt/cache
OMP_NUM_THREADS: 8
MKL_NUM_THREADS: 8
BASE_PATH: benchmark_outputs

jobs:
torch_pipelines_cuda_benchmark_tests:
torch_models_cuda_benchmark_tests:
env:
SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL_BENCHMARK }}
name: Torch Core Pipelines CUDA Benchmarking Tests
name: Torch Core Models CUDA Benchmarking Tests
strategy:
fail-fast: false
max-parallel: 1
runs-on:
group: aws-g6-4xlarge-plus
group: aws-g6e-4xlarge
container:
image: diffusers/diffusers-pytorch-compile-cuda
options: --shm-size "16gb" --ipc host --gpus 0
image: diffusers/diffusers-pytorch-cuda
options: --shm-size "16gb" --ipc host --gpus all
steps:
- name: Checkout diffusers
uses: actions/checkout@v3
uses: actions/checkout@v6
with:
fetch-depth: 2
- name: NVIDIA-SMI
run: |
nvidia-smi
- name: Install dependencies
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install pandas peft
python -m uv pip uninstall transformers && python -m uv pip install transformers==4.48.0
apt update
apt install -y libpq-dev postgresql-client
uv pip install -e ".[quality]"
uv pip install -r benchmarks/requirements.txt
- name: Environment
run: |
python utils/print_env.py
- name: Diffusers Benchmarking
env:
HF_TOKEN: ${{ secrets.DIFFUSERS_BOT_TOKEN }}
BASE_PATH: benchmark_outputs
HF_TOKEN: ${{ secrets.DIFFUSERS_HF_HUB_READ_TOKEN }}
run: |
export TOTAL_GPU_MEMORY=$(python -c "import torch; print(torch.cuda.get_device_properties(0).total_memory / (1024**3))")
cd benchmarks && mkdir ${BASE_PATH} && python run_all.py && python push_results.py
cd benchmarks && python run_all.py

- name: Push results to the Hub
env:
HF_TOKEN: ${{ secrets.DIFFUSERS_BOT_TOKEN }}
run: |
cd benchmarks && python push_results.py
mkdir $BASE_PATH && cp *.csv $BASE_PATH

- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v4
uses: actions/upload-artifact@v6
with:
name: benchmark_test_reports
path: benchmarks/benchmark_outputs
path: benchmarks/${{ env.BASE_PATH }}

- name: Report success status
if: ${{ success() }}
Expand Down
57 changes: 40 additions & 17 deletions .github/workflows/build_docker_images.yml
Original file line number Diff line number Diff line change
Expand Up @@ -25,10 +25,10 @@ jobs:
if: github.event_name == 'pull_request'
steps:
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v1
uses: docker/setup-buildx-action@v3

- name: Check out code
uses: actions/checkout@v3
uses: actions/checkout@v6

- name: Find Changed Dockerfiles
id: file_changes
Expand All @@ -38,15 +38,43 @@ jobs:
token: ${{ secrets.GITHUB_TOKEN }}

- name: Build Changed Docker Images
env:
CHANGED_FILES: ${{ steps.file_changes.outputs.all }}
run: |
CHANGED_FILES="${{ steps.file_changes.outputs.all }}"
echo "$CHANGED_FILES"
ALLOWED_IMAGES=(
diffusers-pytorch-cpu
diffusers-pytorch-cuda
diffusers-pytorch-xformers-cuda
diffusers-pytorch-minimum-cuda
diffusers-doc-builder
)

declare -A IMAGES_TO_BUILD=()

for FILE in $CHANGED_FILES; do
if [[ "$FILE" == docker/*Dockerfile ]]; then
DOCKER_PATH="${FILE%/Dockerfile}"
DOCKER_TAG=$(basename "$DOCKER_PATH")
echo "Building Docker image for $DOCKER_TAG"
docker build -t "$DOCKER_TAG" "$DOCKER_PATH"
# skip anything that isn't still on disk
if [[ ! -e "$FILE" ]]; then
echo "Skipping removed file $FILE"
continue
fi

for IMAGE in "${ALLOWED_IMAGES[@]}"; do
if [[ "$FILE" == docker/${IMAGE}/* ]]; then
IMAGES_TO_BUILD["$IMAGE"]=1
fi
done
done

if [[ ${#IMAGES_TO_BUILD[@]} -eq 0 ]]; then
echo "No relevant Docker changes detected."
exit 0
fi

for IMAGE in "${!IMAGES_TO_BUILD[@]}"; do
DOCKER_PATH="docker/${IMAGE}"
echo "Building Docker image for $IMAGE"
docker build -t "$IMAGE" "$DOCKER_PATH"
done
if: steps.file_changes.outputs.all != ''

Expand All @@ -65,27 +93,22 @@ jobs:
image-name:
- diffusers-pytorch-cpu
- diffusers-pytorch-cuda
- diffusers-pytorch-compile-cuda
- diffusers-pytorch-xformers-cuda
- diffusers-pytorch-minimum-cuda
- diffusers-flax-cpu
- diffusers-flax-tpu
- diffusers-onnxruntime-cpu
- diffusers-onnxruntime-cuda
- diffusers-doc-builder

steps:
- name: Checkout repository
uses: actions/checkout@v3
uses: actions/checkout@v6
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v1
uses: docker/setup-buildx-action@v3
- name: Login to Docker Hub
uses: docker/login-action@v2
uses: docker/login-action@v3
with:
username: ${{ env.REGISTRY }}
password: ${{ secrets.DOCKERHUB_TOKEN }}
- name: Build and push
uses: docker/build-push-action@v3
uses: docker/build-push-action@v6
with:
no-cache: true
context: ./docker/${{ matrix.image-name }}
Expand Down
26 changes: 26 additions & 0 deletions .github/workflows/build_pr_documentation.yml
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,33 @@ concurrency:
cancel-in-progress: true

jobs:
check-links:
runs-on: ubuntu-latest

steps:
- name: Checkout repository
uses: actions/checkout@v6

- name: Set up Python
uses: actions/setup-python@v6
with:
python-version: '3.10'

- name: Install uv
run: |
curl -LsSf https://astral.sh/uv/install.sh | sh
echo "$HOME/.cargo/bin" >> $GITHUB_PATH

- name: Install doc-builder
run: |
uv pip install --system git+https://github.com/huggingface/doc-builder.git@main

- name: Check documentation links
run: |
uv run doc-builder check-links docs/source/en

build:
needs: check-links
uses: huggingface/doc-builder/.github/workflows/build_pr_documentation.yml@main
with:
commit_sha: ${{ github.event.pull_request.head.sha }}
Expand Down
22 changes: 22 additions & 0 deletions .github/workflows/codeql.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,22 @@
---
name: CodeQL Security Analysis For Github Actions

on:
push:
branches: ["main"]
workflow_dispatch:
# pull_request:

jobs:
codeql:
name: CodeQL Analysis
uses: huggingface/security-workflows/.github/workflows/codeql-reusable.yml@v1
permissions:
security-events: write
packages: read
actions: read
contents: read
with:
languages: '["actions","python"]'
queries: 'security-extended,security-and-quality'
runner: 'ubuntu-latest' #optional if need custom runner
35 changes: 19 additions & 16 deletions .github/workflows/mirror_community_pipeline.yml
Original file line number Diff line number Diff line change
Expand Up @@ -24,7 +24,6 @@ jobs:
mirror_community_pipeline:
env:
SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL_COMMUNITY_MIRROR }}

runs-on: ubuntu-22.04
steps:
# Checkout to correct ref
Expand All @@ -39,54 +38,58 @@ jobs:
# If ref is 'refs/heads/main' => set 'main'
# Else it must be a tag => set {tag}
- name: Set checkout_ref and path_in_repo
env:
EVENT_NAME: ${{ github.event_name }}
EVENT_INPUT_REF: ${{ github.event.inputs.ref }}
GITHUB_REF: ${{ github.ref }}
run: |
if [ "${{ github.event_name }}" == "workflow_dispatch" ]; then
if [ -z "${{ github.event.inputs.ref }}" ]; then
if [ "$EVENT_NAME" == "workflow_dispatch" ]; then
if [ -z "$EVENT_INPUT_REF" ]; then
echo "Error: Missing ref input"
exit 1
elif [ "${{ github.event.inputs.ref }}" == "main" ]; then
elif [ "$EVENT_INPUT_REF" == "main" ]; then
echo "CHECKOUT_REF=refs/heads/main" >> $GITHUB_ENV
echo "PATH_IN_REPO=main" >> $GITHUB_ENV
else
echo "CHECKOUT_REF=refs/tags/${{ github.event.inputs.ref }}" >> $GITHUB_ENV
echo "PATH_IN_REPO=${{ github.event.inputs.ref }}" >> $GITHUB_ENV
echo "CHECKOUT_REF=refs/tags/$EVENT_INPUT_REF" >> $GITHUB_ENV
echo "PATH_IN_REPO=$EVENT_INPUT_REF" >> $GITHUB_ENV
fi
elif [ "${{ github.ref }}" == "refs/heads/main" ]; then
echo "CHECKOUT_REF=${{ github.ref }}" >> $GITHUB_ENV
elif [ "$GITHUB_REF" == "refs/heads/main" ]; then
echo "CHECKOUT_REF=$GITHUB_REF" >> $GITHUB_ENV
echo "PATH_IN_REPO=main" >> $GITHUB_ENV
else
# e.g. refs/tags/v0.28.1 -> v0.28.1
echo "CHECKOUT_REF=${{ github.ref }}" >> $GITHUB_ENV
echo "PATH_IN_REPO=$(echo ${{ github.ref }} | sed 's/^refs\/tags\///')" >> $GITHUB_ENV
echo "CHECKOUT_REF=$GITHUB_REF" >> $GITHUB_ENV
echo "PATH_IN_REPO=$(echo $GITHUB_REF | sed 's/^refs\/tags\///')" >> $GITHUB_ENV
fi
- name: Print env vars
run: |
echo "CHECKOUT_REF: ${{ env.CHECKOUT_REF }}"
echo "PATH_IN_REPO: ${{ env.PATH_IN_REPO }}"
- uses: actions/checkout@v3
- uses: actions/checkout@v6
with:
ref: ${{ env.CHECKOUT_REF }}

# Setup + install dependencies
- name: Set up Python
uses: actions/setup-python@v4
uses: actions/setup-python@v6
with:
python-version: "3.10"
- name: Install dependencies
run: |
python -m pip install --upgrade pip
pip install --upgrade pip
pip install --upgrade huggingface_hub

# Check secret is set
- name: whoami
run: huggingface-cli whoami
run: hf auth whoami
env:
HF_TOKEN: ${{ secrets.HF_TOKEN_MIRROR_COMMUNITY_PIPELINES }}

# Push to HF! (under subfolder based on checkout ref)
# https://huggingface.co/datasets/diffusers/community-pipelines-mirror
- name: Mirror community pipeline to HF
run: huggingface-cli upload diffusers/community-pipelines-mirror ./examples/community ${PATH_IN_REPO} --repo-type dataset
run: hf upload diffusers/community-pipelines-mirror ./examples/community ${PATH_IN_REPO} --repo-type dataset
env:
PATH_IN_REPO: ${{ env.PATH_IN_REPO }}
HF_TOKEN: ${{ secrets.HF_TOKEN_MIRROR_COMMUNITY_PIPELINES }}
Expand All @@ -99,4 +102,4 @@ jobs:
- name: Report failure status
if: ${{ failure() }}
run: |
pip install requests && python utils/notify_community_pipelines_mirror.py --status=failure
pip install requests && python utils/notify_community_pipelines_mirror.py --status=failure
Loading