REPA
[ICLR'25 Oral] Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think
Align noisy images with trained models to improve the quality of generated visuals.
Claim its page: indexed whatever its rank, translated into six languages, and enriched with what you write yourself.
Get an email alert on its next release or when it starts trending — never miss the moment.
Free · no card · unsubscribe anytime[ICLR'25 Oral] Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think
REPA has 1.7k stars on GitHub. It has been forked 97 times. REPA is written mainly in Python. It has been in active development since 2024. REPA is available under the MIT license. Its main topics are diffusion, diffusion-models.
What is REPA?
[ICLR'25 Oral] Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think
Is REPA open source?
REPA is an open-source project. It is released under the MIT license.
Is REPA free?
Yes. REPA is free and open source — you can use, modify and self-host it.
What license does REPA use?
REPA is available under the MIT license.
What language is REPA written in?
REPA is written mainly in Python.
Add this live badge to your README — your GitHub stars and directory rank, refreshed daily.
[](https://olud.ai/project/sihyun-yu-repa.html)
Measured from GitHub topics shared by both projects, weighted by how rare each topic is.