diffusers

diffusers copied to clipboard

Add Kohya fix to SD pipeline for high resolution generation

7

# What does this PR do? Adds Kohya fix to Stable Diffusion pipeline. Fixes https://github.com/huggingface/diffusers/issues/7265. # To Test? Here is a minimal example to test the pipeline. You can disable...

sajadn

training example for instruct pix2pix doesn't zero out embeds

8

### Describe the bug When running inference on SDXL, the config specifies to zero out the embedding when the prompt is empty. ### Reproduction ```py # Get null conditioning def...

bghira

Kandinsky 3.1 has updated a few new features(including Kandinsky Flash) a few weeks ago which is yet supported in diffusers. Please check out https://github.com/ai-forever/Kandinsky-3

EYcab

Support for hpcaitech OpenSora's STDiT for text2video and text2image generation

5

### Model/Pipeline/Scheduler description STDiT builds on Latte and DiT and yields a trade-off between generation quality and speed https://github-production-user-asset-6210df.s3.amazonaws.com/99191637/313485495-983a1965-a374-41a7-a76b-c07941a6c1e9.mp4?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAVCODYLSA53PQK4ZA%2F20240318%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20240318T093601Z&X-Amz-Expires=300&X-Amz-Signature=89fc6e69755160d4b0c00efc5a166a04405f29a99f464410dfa53b73e251a0fd&X-Amz-SignedHeaders=host&actor_id=14872007&key_id=0&repo_id=760231710 ### Open source status - [X] The model implementation is available....

kabachuha

Provided pooled_prompt_embeds is overwritten via prompt_embeds[0]

10

https://github.com/huggingface/diffusers/blob/25caf24ef90fc44074f4fd3712f6ed5a1db4a5c3/src/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl.py#L386 Simple fix: ```python pooled_prompt_embeds = prompt_embeds[0] if pooled_prompt_embeds is not None else pooled_prompt_embeds ``` Sorry this isn't a pr :P

cloneofsimo

fix for #7365, prevent pipelines from overriding provided prompt embeds

4

# What does this PR do? Fixes #7365 ## Before submitting - [x] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's...

bghira

Request to implement FreeNoise, a new diffusion scheduler

15

### Model/Pipeline/Scheduler description FreeNoise is a tuning-free and time-efficient paradigm for longer video generation based on pretrained video diffusion models. In other words, it is another *free lunch* that can...

arthur-qiu

Tuning-Free Noise Rectification for High Fidelity Image-to-Video Generation

6

### Model/Pipeline/Scheduler description Applying pretrained Text-to-Video (T2V) Diffusion models to Image-to-video (I2V) generation tasks using SDEdit often results in low source image fidelity in open domains. This method achieves high...

clarencechen

non power of 2 sized images with stable cascade fail

3

### Describe the bug I am not sure if this is a bug, but the documentation does not mention a limitation to power of 2 that I can find. Generating...

Teriks

### Describe the bug I have tried to train a controlnet with lora, but after I merge the controlnet model and lora, the weight of controlnet seems not changed. (maybe...

lavinal712