huggingface/diffusers

Pull Requests Commits

faster model loading on cuda.

sayakpaul committed 1 year ago

51d1436a

[train_dreambooth_lora_sdxl.py] Fix the LR Schedulers when num_train_epochs is passed in a distributed training env (#11240)

kghamilton89 committed 1 year ago

Verified 0dec414d

[Flux LoRAs] fix lr scheduler bug in distributed scenarios (#11242)

linoytsaban committed 1 year ago

Verified 44eeba07

[Wan2.1-FLF2V] update conversion script (#11365)

yiyixuxu committed 1 year ago

Verified 5873377a

update output for Hidream transformer (#11366)

yiyixuxu committed 1 year ago

Verified 5a2e0f71

Fix: `StableDiffusionXLControlNetAdapterInpaintPipeline` incorrectly inherited `StableDiffusionLoraLoaderMixin` (#11357)

Kazuki-Yoda committed 1 year ago

Verified ef47726e

support Wan-FLF2V (#11353)

yiyixuxu committed 1 year ago

Verified 0021bfa1

[BNB] Fix test_moving_to_cpu_throws_warning (#11356)

SunMarc committed 1 year ago

Verified bbd0c161

enable 2 test cases on XPU (#11332)

yao-matrix committed 1 year ago

Verified eef3d659

Update controlnet_flux.py (#11350)

haofanwang committed 1 year ago

Verified ee6ad51d

[bitsandbytes] improve dtype mismatch handling for bnb + lora. (#11270)

sayakpaul committed 1 year ago

Verified 4397f59a

[Hi Dream] follow-up (#11296)

yiyixuxu committed 1 year ago

Verified 05679329

[LoRA] Propagate `hotswap` better (#11333)

sayakpaul committed 1 year ago

Verified 29d2afbf

[docs] add note about use_duck_shape in auraflow docs. (#11348)

sayakpaul committed 1 year ago

Verified b00a564d

[chore] fix lora docs utils (#11338)

sayakpaul committed 1 year ago

Verified efc9d68b

Fix wrong dtype argument name as torch_dtype (#11346)

nPeppon committed 1 year ago

Verified 3e59d531

[BUG] fixed _toctree.yml alphabetical ordering (#11277)

ishan-modi committed 1 year ago

Verified d63e6fcc

Hunyuan I2V fast tests fix (#11341)

DN6 committed 1 year ago

Verified 59f1b7b1

[docs] add a snippet for compilation in the auraflow docs. (#11327)

sayakpaul committed 1 year ago

Verified ce1063ac

[single file] enable telemetry for single file loading when using GGUF. (#11284)

sayakpaul committed 1 year ago

Verified 7212f35d

unpin torch versions for onnx Dockerfile (#11290)

sayakpaul committed 1 year ago

Verified 3252d7ad

Fix Hunyuan I2V for `transformers>4.47.1` (#11293)

DN6 committed 1 year ago

Verified b316104d

another fix for FlowMatchLCMScheduler forgotten import (#11330)

asomoza committed 1 year ago

Verified d3b2699a

post release 0.33.0 (#11255)

sayakpaul committed 1 year ago

Verified 4b868f14

Rewrite AuraFlowPatchEmbed.pe_selection_index_based_on_dim to be torch.compile compatible (#11297)

AstraliteHeart committed 1 year ago

Verified b6156aaf

[docs] fix hidream docstrings. (#11325)

sayakpaul committed 1 year ago

Verified 7ecfe291

fix CPU offloading related fail cases on XPU (#11288)

yao-matrix committed 1 year ago

Verified 7edace9a

Fix vae.Decoder prev_output_channel (#11280)

hlky committed 1 year ago

Verified 6e80d240

[LoRA] Add LoRA support to AuraFlow (#10216)

hameerabbasi committed 1 year ago

Verified 9352a5ca

[docs] Promote `AutoModel` usage (#11300)

sayakpaul committed 1 year ago

Verified cefa28f4

Older