huggingface/diffusers

Pull Requests Commits

SchedulerMixin from_pretrained and ConfigMixin Self type annotation (#11192)

hlky committed 189 days ago

Verified 78c2fdc5

Fix enable_sequential_cpu_offload in CogView4Pipeline (#11195)

hlky committed 189 days ago

Verified 54dac3a8

[docs] `torch_dtype` map (#11194)

hlky committed 189 days ago

Verified e5c6027e

Revert `save_model` in ModelMixin save_pretrained and use safe_serialization=False in test (#11196)

hlky committed 189 days ago

Verified da857beb

[tests] HunyuanDiTControlNetPipeline inference precision issue on XPU (#11197)

faaany committed 189 days ago

Verified 52b460fe

allow models to run with a user-provided dtype map instead of a single dtype (#10301)

hlky committed 190 days ago

Verified d8c617cc

remove unnecessary call to `F.pad` (#10620)

bm-synth committed 190 days ago

Verified fe2b3974

fix: for checking mandatory and optional pipeline components (#11189)

elismasilva committed 190 days ago

Verified be0b7f55

fix autocast (#11190)

jiqing-feng committed 190 days ago

Verified 4d5a96e4

map BACKEND_RESET_MAX_MEMORY_ALLOCATED to reset_peak_memory_stats on XPU (#11191)

yao-matrix committed 190 days ago

Verified a7f07c1e

[WIP] Add Wan Video2Video (#11053)

DN6 committed 190 days ago

Verified df1d7b01

[tests] no hard-coded cuda (#11186)

faaany committed 190 days ago

Verified 5a6edac0

Bug fix in LTXImageToVideoPipeline.prepare_latents() when latents is already set (#10918)

kakukakujirori committed 191 days ago

Verified e8fc8b1f

Add `latents_mean` and `latents_std` to `SDXLLongPromptWeightingPipeline` (#11034)

hlky committed 191 days ago

Verified d6f4774c

[Docs] Fix environment variables in `installation.md` (#11179)

remarkablemark committed 191 days ago

Verified eb50deff

Raise warning and round down if Wan num_frames is not 4k + 1 (#11167)

a-r-r-o-w committed 192 days ago

Verified 2c59af72

Fix LatteTransformer3DModel dtype mismatch with enable_temporal_attentions (#11139)

hlky committed 193 days ago

Verified 75d7e5cc

[Docs] Update Wan Docs with memory optimizations (#11089)

DN6 committed 194 days ago

Verified 617c208b

WanI2V encode_image (#11164)

hlky committed 194 days ago

Verified 5d970a4a

Set self._hf_peft_config_loaded to True when LoRA is loaded using `load_lora_adapter` in PeftAdapterMixin class (#11155)

kentdan3msu committed 196 days ago

Verified de6a88c2

[Quantization] dtype fix for GGUF + fix BnB tests (#11159)

DN6 committed 196 days ago

Verified 7dc52ea7

add a timestep scale for sana-sprint teacher model (#11150)

lawrence-cj committed 197 days ago

Verified 739d6ec7

Improve information about group offloading and layerwise casting (#11101)

a-r-r-o-w committed 198 days ago

Verified 1ddf3f3a

[doc] Fix Korean Controlnet Train doc (#11141)

flyxiv committed 198 days ago

Verified 7aac77af

New HunyuanVideo-I2V (#11066)

a-r-r-o-w committed 198 days ago

Verified 8907a70a

[fix SANA-Sprint] (#11142)

lawrence-cj committed 199 days ago

Verified 5dbe4f5d

Modify the implementation of retrieve_timesteps in CogView4-Control. (#11125)

zRzRzRzRzRzRzR committed 199 days ago

Verified 1d37f420

Update README and example code for AnyText usage (#11028)

tolgacangoz committed 199 days ago

Verified 0213179b

Don't override `torch_dtype` and don't use when `quantization_config` is set (#11039)

hlky committed 201 days ago

Verified a7d53a59

add sana-sprint (#11074)

yiyixuxu committed 201 days ago

Verified 8a63aa5e

Newer Older