huggingface/diffusers

Pull Requests Commits

Merge branch 'main' into hidream-non-diffusers-lora

sayakpaul committed 1 year ago

Verified cc033515

make fix-copies

sayakpaul committed 1 year ago

e1968390

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

sayakpaul committed 1 year ago

Verified 7acf8345

support non-diffusers hidream loras

sayakpaul committed 1 year ago

3c84a3c1

feat: pipeline-level quantization config (#11130)

sayakpaul committed 1 year ago

Verified 599c8871

[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

enable lora cases on XPU (#11506)

yao-matrix committed 1 year ago

Verified 8c661ea5

Hunyuan Video Framepack (#11428)

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

Verified d7ffe601

[LoRA] use `removeprefix` to preserve sanity. (#11493)

sayakpaul committed 1 year ago

Verified 10bee525

update dep table. (#11504)

sayakpaul committed 1 year ago

Verified d88ae1f5

Update setup.py to pin min version of `peft` (#11502)

sayakpaul committed 1 year ago

Verified 53f1043c

Fix torchao docs typo for fp8 granular quantization (#11473)

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

Verified 1fa56394

Update training script for txt to img sdxl with lora supp with new interpolation. (#11496)

RogerSinghChugh committed 1 year ago

Verified ed4efbd6

Set LANCZOS as the default interpolation method for image resizing. (#11492)

yijun-lee committed 1 year ago

Verified 9c29e938

[training] feat: enable quantization for hidream lora training. (#11494)

sayakpaul committed 1 year ago

Verified 071807c8

[train_dreambooth_lora_lumina2] Add LANCZOS as the default interpolation mode for image resizing (#11491)

cjfghk5697 committed 1 year ago

Verified ee1516e5

[train_dreambooth_lora_sdxl] Add --image_interpolation_mode option for image resizing (default to lanczos) (#11490)

MinJu-Ha committed 1 year ago

Verified ec932399

[train_text_to_image_sdxl]Add LANCZOS as default interpolation mode for image resizing (#11455)

ParagEkbote committed 1 year ago

Verified fc5e9066

[Feature] Implement tiled VAE encoding/decoding for Wan model. (#11414)

c8ef committed 1 year ago

Verified 8520d496

enable semantic diffusion and stable diffusion panorama cases on XPU (#11459)

yao-matrix committed 1 year ago

Verified a674914f

[train_dreambooth_lora_flux_advanced] Add LANCZOS as the default interpolation mode for image resizing (#11472)

ysurs committed 1 year ago

Verified ec3d5828

Older