diffusers

diffusers copied to clipboard

Fix saving text encoder weights and kohya weights in advanced dreambooth lora script

1

# What does this PR do? There seems to be an issue with 1. Saving Kohya style weights in the dreambooth advanced lora script. The save path can output a...

DN6

[WIP] Stable Audio integration

4

# What does this PR do? Stability AI recently open-sourced [Stable Audio 1.0](https://huggingface.co/stabilityai/stable-audio-open-1.0), which can be run using their [toolkit library](https://github.com/Stability-AI/stable-audio-tools/) . Contrarily to most diffusion models, the diffusion process...

ylacombe

SD3 cannot finetunes a better model (hand and face deformation)?

3

### Describe the bug I want to finetune sd3 to improve its human generation quality with 3million high-quality human datasets (which has been proven useful on sdxl and other models)....

KaiWU5

set `is_sequential_cpu_offload = True` only when some component is on cpu and has AlignDevicesHook simultaneously

3

# What does this PR do? this pr is talking about when the is_sequential_cpu_offload should be set to True. before we got device_map feature for pipline, if any component in...

zhangvia

Batched Text to image generation really slow with pipeline

7

### Describe the bug batched diffusers pipeline inference is really slow ### Reproduction ``` class prompt_data(Dataset): def __init__(self,prompt_file): self.txt_prompts = [] with open(prompt_file,'r') as file:self.txt_prompts = file.readlines() def __len__(self): return...

nighting0le01

Add PAG support to SD1.5

9

We recently integrated PAG into diffusers! See this PR [here] (https://github.com/huggingface/diffusers/pull/7944) we added PAG to SDXL we also want to add PAG support to SD1.5 pipelines! we will need: -...

yiyixuxu

Hello @manuelbrack, I was reviewing the source code and came across a line that seems to have a hardcoded value in a conditional statement. The line in question is: https://github.com/huggingface/diffusers/blob/0bae6e447cba0459456c4f7e7e87d7db141d3235/src/diffusers/pipelines/ledits_pp/pipeline_leditspp_stable_diffusion.py#L1053...

ardofski

[BUG] convert T5 text encoder to float16 results corrupted image

13

### Describe the bug I have tested PixArt-Sigma with following code, where I load text_encoder separately since I will fine-tune it in later. I found `T5EncoderModel.from_pretrained(torch_dtype=torch.float16)` is very different from...

Luciennnnnnn

Using `torchsde.BrownianInterval` instead of `torchsde.BrownianTree` in class `BatchedBrownianTree`

2

**Is your feature request related to a problem? Please describe.** When I was doing some optimization for my pipeline, i found that the BrownianTree somehow took a bit more time....

dianyo

add conv_in_channels in def save_motion_modules

1

# What does this PR do? In [save_motion_modules](https://github.com/huggingface/diffusers/blob/7ebd359446c2cb31bfbbbd98046cd916de8bdc7b/src/diffusers/models/unets/unet_motion_model.py#L592), current implementation doesn't pass `conv_in_channels` to the MotionAdapter constructor. This would incorrectly result into `"conv_in_channels": null` in the saved `config.json`, disregarding the...

ernestchu