facebookresearch/DiT

Health v2 · maintenance only

Flags: no_releases archived no_license

Adoption not part of the score

8689 stars · 810 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Official PyTorch implementation of Diffusion Transformers (DiT) from the paper 'Scalable Diffusion Models with Transformers', including model definitions, pre-trained ImageNet weights, and training/sampling scripts. It replaces the U-Net backbone of latent diffusion models with a transformer operating on latent patches.

Use cases

• generate class-conditional images with a diffusion transformer
• run pre-trained DiT-XL/2 models on ImageNet 256x256 or 512x512
• train a latent diffusion model with a transformer backbone instead of U-Net
• experiment with scalability of diffusion models measured in Gflops
• reproduce state-of-the-art FID results on class-conditional ImageNet generation

When to choose

• you want the reference implementation of DiT with official pre-trained weights
• you need a simple, self-contained PyTorch codebase for diffusion transformer research
• you want to build on the architecture used by later models like Stable Diffusion 3

When to avoid

• you need a production-ready text-to-image pipeline with broad feature support
• you want a maintained general-purpose diffusion library like Hugging Face diffusers
• you need models trained on datasets other than ImageNet class-conditional generation

Facets

library · maturity maintenance

machine-learning deep-learning image-processing deep-learning machine-learning image-processing artificial-intelligence python diffusion-models transformers pytorch image-generation research-code pretrained-models gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="facebookresearch/DiT")