About
Stable Diffusion is a family of open text-to-image generation models developed by Stability AI. Successive generations — from latent diffusion models to Stable Diffusion 3 and 3.5 based on the Multimodal Diffusion Transformer (MMDiT) architecture and rectified-flow / flow-matching training — are released with open weights under the Stability AI Community License. The models can be run locally as well as via APIs and partner platforms.
