Zubnet AILearnWiki › Diffusion Transformer
Models

Diffusion Transformer

DiT
An architecture that replaces the U-Net backbone traditionally used in diffusion models with a Transformer. DiT applies the attention mechanism to image generation, enabling the same scaling behavior that made LLMs so powerful. Sora, Flux, Stable Diffusion 3, and most state-of-the-art image and video generators use DiT or variants.

Why it matters

DiT unified the worlds of language and image generation under a single architectural paradigm: the Transformer. This means the scaling laws, training techniques, and optimization strategies developed for LLMs largely transfer to image and video generation. It's why image quality has improved so rapidly — the field is riding the same scaling curve as language.

Deep Dive

The original DiT paper (Peebles & Xie, 2023) showed that simply replacing the U-Net with a standard Transformer and scaling it up produced better image quality. The Transformer processes image patches (similar to Vision Transformers) with added conditioning from the diffusion timestep and class labels. The key finding: DiT follows clear scaling laws — larger models and more compute produce predictably better images, just like with LLMs.

From U-Net to Transformer

U-Nets process images at multiple resolutions, downsampling then upsampling with skip connections. This inductive bias was useful when compute was limited, but it introduces architectural complexity and doesn't scale as cleanly. Transformers, with their uniform architecture, are simpler to scale and benefit more from additional compute and data. The trade-off: Transformers are more memory-hungry due to the quadratic attention over all image patches.

MM-DiT: Multi-Modal DiT

Stable Diffusion 3 and Flux use MM-DiT (Multi-Modal DiT), which processes text and image tokens through separate streams that interact via cross-attention. This is more effective than the simpler text-conditioning used in the original DiT. The text stream uses a frozen text encoder (like T5 or CLIP), and the image stream uses the diffusion process. The two streams exchange information at each Transformer block.

Related Concepts

← All Terms
← Diffusion Model Distillation →