Diffusion & generation

DiT block (diffusion transformer)

A transformer block operating on image-latent tokens, modulated by the diffusion timestep.

First page of Scalable Diffusion Models with TransformersIntroduced inDec 2022Scalable Diffusion Models with TransformersPeebles & Xie · arXiv 2212.09748 ↗
DiT block: a transformer steered by timelatent tokens pass through attention and MLP while the timestep steers each normattentionMLPt + promptscale + shift

Diffusion transformers replace the classic U-Net with a stack of transformer blocks over patchified latent tokens. Each block is conditioned on the timestep (and prompt) through adaptive layer-norm modulation. The timestep embedding shifts and scales activations inside every block. DiTs scale more predictably than U-Nets and power most recent image and video generators.

Adoption over time

Share of new models that have included DiT blocks over time.

9%202220232024202520269%20222023202420252026

See it in real models

Open any of these on hfviewer to find this block in the interactive architecture graph.

Comfy-Org/MiniMax-H3 architecture graphComfy-Org/MiniMax-H3↓ 23.3M · ♡ 2kOpen in visualizer Comfy-Org/Krea-2 architecture graphComfy-Org/Krea-2↓ 9.8M · ♡ 573Open in visualizer coqui/XTTS-v2 architecture graphcoqui/XTTS-v2text-to-speech · ↓ 7.2M · ♡ 4kOpen in visualizer MiniMaxAI/MiniMax-H3 architecture graphMiniMaxAI/MiniMax-H3image-text-to-video · ↓ 4.1M · ♡ 6kOpen in visualizer stabilityai/stable-diffusion-xl-base-1.0 architecture graphstabilityai/stable-diffusion-xl-base-1.0text-to-image · ↓ 3.1M · ♡ 8kOpen in visualizer ResembleAI/chatterbox architecture graphResembleAI/chatterboxtext-to-speech · ↓ 1.8M · ♡ 2kOpen in visualizer stable-diffusion-v1-5/stable-diffusion-v1-5 architecture graphstable-diffusion-v1-5/stable-diffusion-v1-5text-to-image · ↓ 1.8M · ♡ 1kOpen in visualizer Lightricks/LTX-2.5 architecture graphLightricks/LTX-2.5image-to-video · ↓ 1.6M · ♡ 5kOpen in visualizer nvidia/Cosmos3-Edge architecture graphnvidia/Cosmos3-Edge↓ 1.5M · ♡ 208Open in visualizer Lightricks/LTX-2.3 architecture graphLightricks/LTX-2.3image-to-video · ↓ 1.1M · ♡ 2kOpen in visualizer unsloth/MiniMax-H3-GGUF architecture graphunsloth/MiniMax-H3-GGUFimage-text-to-video · ↓ 979k · ♡ 302Open in visualizer stabilityai/sdxl-turbo architecture graphstabilityai/sdxl-turbotext-to-image · ↓ 886k · ♡ 3kOpen in visualizer

View 420 catalog matches for DiT block (diffusion transformer) →

Related concepts

VAE (latent autoencoder) Denoising loop / scheduler Timestep embedding

hfviewer renders the full architecture of 6,700+ Hugging Face models as interactive graphs. Hover any block to see what it does, with this model’s real numbers.

Browse all model graphs →