Diffusion models generate images by starting with random noise and progressively denoising it into a coherent image. They power tools like Stable Diffusion, DALL-E, and Midjourney. The process involves a forward pass (adding noise) and a reverse pass (removing noise). They produce some of the highest-quality AI-generated images available today.








