TAESD (Tiny AutoEncoder for Stable Diffusion) is a very small model that shares the same latent space as Stable Diffusion's VAE. It enables fast, real-time previewing of the image generation process inside tools like diffusers. The repository provides safetensors weights for easy integration.
Taesd is an Image generation project. It focuses on previewing Stable Diffusion image generation in real time without running the full VAE. It is built as an open-source project for developers. The project is open source (MIT). Taesd is available on the web, the command line, and API.
Behind Taesd is madebyollin, and it first shipped in 2023. The project is developed in the open on GitHub with 954 stars. Key capabilities include Tiny Autoencoder, Real-time Preview, and Stable Diffusion Compatible.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match