Diffusion Model

Models and Architectures

The AI technique behind most image generators - it creates pictures by gradually turning random noise into a coherent image.

A diffusion model learns to reverse a corruption process. During training it watches images being progressively destroyed by noise, and learns to undo each step. To generate a new image it starts from pure noise and denoises step by step toward something that matches your prompt.Midjourney, Stable Diffusion, and DALL-E all rely on this technique, and it is spreading to video and audio generation as well.