huggingface/diffusers

Pull Requests Commits

[tests] Enable testing for HiDream transformer (#11478)

sayakpaul committed 1 year ago

Verified b555a037

LTX Video 0.9.7 (#11516)

a-r-r-o-w committed 1 year ago

Verified 06fee551

[LoRA] small change to support Hunyuan LoRA Loading for FramePack (#11546)

linoytsaban committed 1 year ago

Verified 8b99f7e1

[train_dreambooth.py] Fix the LR Schedulers when num_train_epochs is passed in a distributed training env (#11239)

kghamilton89 committed 1 year ago

Verified 07dd6f8c

fix: remove `torch_dtype="auto"` option from docstrings (#11513)

johannaSommer committed 1 year ago

Verified f8d4a1e2

Fix typo in train_diffusion_orpo_sdxl_lora_wds.py (#11541)

Meeex2 committed 1 year ago

Verified ddd0cfb4

Add VisualCloze (#11377)

lzyhha committed 1 year ago

Verified 4f438de3

[test_models_transformer_ltx.py] help us test torch.compile() for impactful models (#11512)

cjfghk5697 committed 1 year ago

Verified 98cc6d05

enable several pipeline integration tests on XPU (#11526)

yao-matrix committed 1 year ago

Verified c3726153

Hunyuan Video Framepack F1 (#11534)

a-r-r-o-w committed 1 year ago

Verified e48f6aee

[tests] add tests for framepack transformer model. (#11520)

sayakpaul committed 1 year ago

Verified 01abfc87

Change Framepack transformer layer initialization order (#11535)

a-r-r-o-w committed 1 year ago

Verified 92fe689f

enable print_env on xpu (#11507)

yao-matrix committed 1 year ago

Verified 0ba1f76d

enable dit integration cases on xpu (#11523)

yao-matrix committed 1 year ago

Verified d6bf268a

[LTXPipeline] Update latents dtype to match VAE dtype (#11533)

james-p-xu committed 1 year ago

Verified 3c0a0129

enable 7 cases on XPU (#11503)

yao-matrix committed 1 year ago

Verified 2d380895

[LoRA] support non-diffusers hidream loras (#11532)

sayakpaul committed 1 year ago

Verified 0c47c954

[Tests] Enable more general testing for `torch.compile()` with LoRA hotswapping (#11322)

sayakpaul committed 1 year ago

Verified 7acf8345

feat: pipeline-level quantization config (#11130)

sayakpaul committed 1 year ago

Verified 599c8871

Merge branch 'main' into repr-quant-config

sayakpaul committed 1 year ago

Verified 895428be

[tests] fix audioldm2 for transformers main. (#11522)

sayakpaul committed 1 year ago

Verified 393aefcd

Conditionally import torchvision in Cosmos transformer (#11524)

a-r-r-o-w committed 1 year ago

Verified 6674a515

Add cross attention type for Sana-Sprint training in diffusers. (#11514)

scxue committed 1 year ago

Verified 784db0ea

[LoRA] make lora alpha and dropout configurable (#11467)

linoytsaban committed 1 year ago

Verified 66e50d4e

Revert "fix audioldm"

sayakpaul committed 1 year ago

c5c34a45

sayakpaul committed 1 year ago

87e508f1

clean up the __Init__ for stable_diffusion (#11500)

yiyixuxu committed 1 year ago

Verified 53bd367b

Cosmos (#10660)

a-r-r-o-w committed 1 year ago

Verified 7b904941

[docs] minor updates to bitsandbytes docs. (#11509)

sayakpaul committed 1 year ago

Verified fb29132b

[lora_conversion] Enhance key handling for OneTrainer components in LORA conversion utility (#11441) (#11487)

iamwavecut committed 1 year ago

Verified 79371661

Newer Older