Diffusion Explainer (Visualizing How Image Generation Works)
A visualization that walks step by step through how Stable Diffusion turns text into an image. On one screen you can follow how the prompt is split into tokens, encoded into vectors, refined from random noise over many timesteps, and finally upscaled.
How to try it
Pick one of the prepared prompts with "Select another prompt", then click a stage box — text representation generation, image representation refining, upscaling — to expand it and see the explanation and intermediate results. Move the timestep to watch the noise being removed, and change the guidance scale to compare how closely the image follows the prompt. Prompts can only be chosen from the prepared list, and the interface is in English only.