Skip to content

Diffusion Model

A generative model that creates images by gradually denoising random noise, powering tools like Stable Diffusion.

It learns to reverse a noising process step by step, so a text prompt steers the denoising toward a matching image. The same idea now extends to video and audio generation.

Related resources

Stable Diffusion WebUI
The most popular community-built graphical interface for Stable Diffusion, developed by AUTOMATIC1111. It provides txt2img, img2img, inpainting and upscaling in a single browser tab, backed by a massive ecosystem of models, LoRAs and extensions. It is the go-to choice for local image generation — provided you have a capable GPU. Aimed at creators who want full control and privacy.
ComfyUI
A node-based, visual workflow interface for image generation models like Stable Diffusion and SDXL. It gives you fine-grained control over every step of the pipeline — from sampling to decoding — and lets you save, share and reuse workflows as files. Best suited for power users who need maximum flexibility and reproducible, complex image generation pipelines.