A diffusion model is trained to denoise: shown a progressively noisier image, it learns to recover the cleaner one. Run backwards from pure static, guided by a text prompt, that skill becomes image generation — the model 'finds' a picture inside the noise that matches the description.
Because they run locally and openly, diffusion models are the practical backbone of commercial image work: style-consistent illustrations, product mockups, inpainting and repair. Control mechanisms — depth maps, edge guidance, reference images — are what turned 'pretty picture' into 'usable asset'.
Related terms
Stable Diffusion
A family of open-weight image models that generate a picture by repeatedly removing noise from a random start, guided by a text description.
ComfyUI
A node-based interface for image generation, where the pipeline is an explicit graph rather than a text box with hidden defaults.
Multimodal model
A model that takes and produces more than text — images, audio, documents — so 'what is wrong in this photo' or 'read this invoice' is one call.
The bench this belongs to
Generative AIStable Diffusion and ComfyUI wired into the place where your content actually gets made, with a model fine-tuned so everything comes out looking like you.
