What it is
Stable Diffusion is Stability AI's family of image models. The current Stable Diffusion 3.5 line comes in Large, Large Turbo, which generates in four steps, and Medium, which is designed to run on consumer hardware; older models such as SDXL Turbo are still offered. You can download the weights and run them in tools like ComfyUI, call them through Stability's Platform API with editing, upscaling and relighting endpoints, or use them through cloud partners.
It suits technical users who want control over their own setup. The Community License is free for research, personal use and organizations with under $1,000,000 in annual revenue; larger companies need a paid license. Running locally needs a capable GPU, and the weights are not under an OSI open-source license.
Best for
- Running an image model on your own GPU
- Fine-tuning and building custom image pipelines
- Developers who need an image API with editing tools