AIDC-AI/Diffusion-SDPO

Diffusion-SDPO: Safeguarded Direct Preference Optimization for Diffusion Models

29
/ 100
Experimental

This project helps AI researchers and machine learning engineers fine-tune diffusion models to better align with specific preferences for generated images. It takes an existing diffusion model and a dataset of preferred and rejected image pairs, then outputs an improved model that generates images more closely matching the desired style or content, without losing the quality of previously preferred outputs.

Use this if you are a machine learning engineer or researcher looking to improve the image generation quality of your diffusion models based on human preferences, ensuring that desired image characteristics are preserved during optimization.

Not ideal if you are an artist or designer looking for a user-friendly application to simply generate images, as this tool requires deep technical knowledge of model training.

AI research Generative AI Diffusion models Model fine-tuning Image generation
No Package No Dependents
Maintenance 6 / 25
Adoption 6 / 25
Maturity 13 / 25
Community 4 / 25

How are scores calculated?

Stars

21

Forks

1

Language

Python

License

Apache-2.0

Last pushed

Nov 11, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/AIDC-AI/Diffusion-SDPO"

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