milmor/diffusion-transformer

Implementation of Diffusion Transformer Model in Pytorch

43
/ 100
Emerging

Replaces standard attention with Linformer to reduce computational complexity, enabling training on single-GPU setups. Implements DiT blocks with adaptive layer normalization (adaLN-Zero) and uses the EDM sampler for improved diffusion scheduling. Includes FID evaluation metrics for image generation quality assessment on ImageNet-64, with plans for classifier-free guidance and latent diffusion variants.

No Package No Dependents
Maintenance 10 / 25
Adoption 9 / 25
Maturity 9 / 25
Community 15 / 25

How are scores calculated?

Stars

73

Forks

11

Language

Python

License

MIT

Last pushed

Jan 06, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/milmor/diffusion-transformer"

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