Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.comfy.org/llms.txt

Use this file to discover all available pages before exploring further.

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Overview

This node allows you to manually select which GPU device the VAE model should be placed on. By default, the VAE is placed on the device assigned by the model loader, but you can pin it to a specific GPU (e.g., gpu:0, gpu:1). If the selected device is not available on your machine, the node will pass the VAE through unchanged and log a message instead of failing.

Inputs

ParameterData TypeRequiredRangeDescription
vaeVAEYesThe VAE model to assign to a specific device.
deviceCOMBOYes"default"
"gpu:0"
"gpu:1"
"gpu:2"
"gpu:3"
"gpu:4"
"gpu:5"
"gpu:6"
"gpu:7"
The target device for the VAE. "default" restores the device assigned by the loader. "gpu:N" pins the VAE to the Nth available GPU. CPU is not a supported choice and will be ignored if provided. (default: "default")

Outputs

Output NameData TypeDescription
vaeVAEThe VAE model, now assigned to the selected device. If the requested device is unavailable or invalid, the VAE is passed through unchanged.

Source fingerprint (SHA-256): 011154043fc02f930b0074de656bb24baf4dfe74bcfd2e89ea76284f0a5b7d8e