A type of generative model that creates images (or video, audio) by starting with pure noise and gradually removing it until a coherent output appears. The model learns to reverse the process of adding noise to real data. Stable Diffusion, DALL-E 3, and Midjourney all use variants of this approach.
Why it matters
Diffusion models dethroned GANs as the dominant image generation technique around 2022. They produce more diverse, controllable outputs and are the backbone of almost every image and video AI tool today.