gudaochangsheng/WaDi

[CVPR 2026] Official PyTorch implementation of WaDi: Weight Direction-aware Distillation for One-step Image Synthesis

21
/ 100
Experimental

This project helps AI researchers and practitioners accelerate image generation from complex models like Stable Diffusion. It takes an existing multi-step image generation model and distills it into a one-step version. The output is a new, faster model that produces high-quality images much more quickly. AI developers and machine learning engineers working on image synthesis will find this useful.

Use this if you need to generate high-quality images instantly from text or other inputs, and want to make your existing diffusion models significantly faster.

Not ideal if you are looking for a completely new image generation model rather than an acceleration method for existing ones, or if you don't work with diffusion models.

AI image generation diffusion models model acceleration real-time image synthesis deep learning optimization
No License No Package No Dependents
Maintenance 13 / 25
Adoption 5 / 25
Maturity 3 / 25
Community 0 / 25

How are scores calculated?

Stars

9

Forks

Language

Python

License

Last pushed

Mar 24, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/gudaochangsheng/WaDi"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.