VAE (Variational Autoencoder)
The neural encoder/decoder that converts between RGB pixel images and compressed diffusion latents.
Updated Oct 17, 2026 · 1 min read
- Term
- VAE (Variational Autoencoder)
- In Short
- The neural encoder/decoder that converts between RGB pixel images and compressed diffusion latents.
- Applies To
- ComfyUI SD 1.5, SDXL, Flux & Video
Definition: VAE (Variational Autoencoder)
The neural encoder/decoder that converts between RGB pixel images and compressed diffusion latents.
A mismatched or missing VAE causes washed-out gray colors or black outputs. SDXL uses sdxl_vae.safetensors, while Flux uses ae.safetensors.
How to tune this in your ComfyUI workflow
Every parameter in ComfyUI is exposed directly on its node rather than hidden behind global settings tabs. You can inspect or override this behavior per pass in ComfyUI Studio.
New to node graphs? Grab the Windows installer at /download and open a built-in template from the workflow menu.
FAQs
What does VAE (Variational Autoencoder) do in ComfyUI?
The neural encoder/decoder that converts between RGB pixel images and compressed diffusion latents. A mismatched or missing VAE causes washed-out gray colors or black outputs. SDXL uses sdxl_vae.safetensors, while Flux uses ae.safetensors.
Where do I adjust this setting on the canvas?
Check the corresponding loader or KSampler widget in your active workflow graph.
Run on your hardware. Free forever.
The full ComfyUI engine: open source, fast, extensible, and yours to run however you want. Download the desktop app and open this workflow on your own machine.
