FreeU: Free Lunch in Diffusion U-Net

[tin2tin]

Model/Pipeline/Scheduler description

FreeU, a method that substantially improves diffusion model sample quality at no costs: no training, no additional parameter introduced, and no increase in memory or sampling time.

Promising results on image and video generation tasks demonstrate that our FreeU can be readily integrated to existing diffusion models, e.g., Stable Diffusion, DreamBooth, ModelScope, Rerender and ReVersion, to improve the generation quality with only a few lines of code. All you need is to adjust two scaling factors during inference.

2023-09-23_xiaohuggg_1705418820796723387_og32HqpAfnM9fzVl.mp4

The code is in the readme here:
https://github.com/ChenyangSi/FreeU

Open source status

• The model implementation is available
• The model weights are available (Only relevant if addition is not a scheduler).

Provide useful links for the implementation

https://chenyangsi.top/FreeU/
https://github.com/ChenyangSi/FreeU

[tin2tin]

Camenduru did a quick fork here with the implementation for SD 1.5 here: Stability-AI/stablediffusion@main...camenduru:stablediffusion:FreeU

By ljleb for auto1111: https://github.com/ljleb/sd-webui-freeu

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.