After the AI generates the image in this small space, the VAE (specifically the decoder) translates it back into a high-resolution image.
It shrinks large images into a smaller "latent" format so the AI can process them more efficiently.
High-quality VAEs, like those included in specialized .zip files, are essential for fixing "washed out" or gray colors and improving the rendering of complex details like eyes and hands. Installation & Use SF_EBL_1.1_vae.zip
Custom models often require a specific VAE to achieve their intended aesthetic. Without it, your generated images might appear:
To use this specific VAE in a tool like , follow these steps: After the AI generates the image in this
Place the resulting .safetensors or .pt file into the models/VAE folder of your Stable Diffusion directory.
Unzip the contents using a tool like 7-Zip or WinRAR . Installation & Use Custom models often require a
In your WebUI, go to the Settings tab, find the Stable Diffusion section, and select the VAE from the SD VAE dropdown menu. Why this specific file?
Please confirm you want to block this member.
You will no longer be able to:
Please allow a few minutes for this process to complete.