---
title: "VAE (Variational Autoencoder) — ComfyUI Glossary & Node Definition"
description: "The neural encoder/decoder that converts between RGB pixel images and compressed diffusion latents. Learn how VAE (Variational Autoencoder) works inside ComfyUI node workflows on Windows."
canonical: https://comfy.ink/words/vae-variational-autoencoder
updated: 2026-10-17
---

# VAE (Variational Autoencoder)

The neural encoder/decoder that converts between RGB pixel images and compressed diffusion latents.

## Facts (2026-10-17)

- **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](https://comfy.ink/studio).

New to node graphs? Grab the Windows installer at [/download](https://comfy.ink/download) and open a built-in template from the workflow menu.

## FAQ

### 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](https://comfy.ink/nodes/ksampler) widget in your active workflow graph.

## Related

- [ComfyUI Glossary](https://comfy.ink/words)
- [KSampler & KSampler (Advanced)](https://comfy.ink/nodes/ksampler)
- [ComfyUI Studio Workspace](https://comfy.ink/studio)
- [ComfyUI VRAM Requirements for SDXL, Flux & Video (6GB–24GB)](https://comfy.ink/guides/comfyui-vram-requirements-guide)
- [Latent Space & LATENT Pin](https://comfy.ink/words/latent-space)

## Download

Free desktop app: [Download Comfy](https://comfy.ink/download)

Markdown mirror: https://comfy.ink/words/vae-variational-autoencoder.md