SD_4XUpscale_Conditioning - ComfyUI Built-in Node Documentation

The SD_4XUpscale_Conditioning node prepares conditioning data for upscaling images using diffusion models. It takes input images and conditioning data, then applies scaling and noise augmentation to create modified conditioning that guides the upscaling process. The node outputs both positive and negative conditioning along with latent representations for the upscaled dimensions.

Inputs

Parameter	Description	Data Type	Required	Range
`images`	Input images to be upscaled	IMAGE	Yes	-
`positive`	Positive conditioning data that guides the generation toward desired content	CONDITIONING	Yes	-
`negative`	Negative conditioning data that steers the generation away from unwanted content	CONDITIONING	Yes	-
`scale_ratio`	Scaling factor applied to the input images (default: 4.0)	FLOAT	No	0.0 - 10.0
`noise_augmentation`	Amount of noise to add during the upscaling process (default: 0.0)	FLOAT	No	0.0 - 1.0

Note: noise_augmentation is an advanced parameter, shown in the node interface under the “Advanced” toggle.

Outputs

Output Name	Description	Data Type
`positive`	Modified positive conditioning with upscaling information applied	CONDITIONING
`negative`	Modified negative conditioning with upscaling information applied	CONDITIONING
`latent`	Empty latent representation matching the upscaled dimensions	LATENT

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

Source fingerprint (SHA-256): f215e890bd86f42d4da9c6f575fc92e65844e2e2056c5610310d8089e5d61902

​Inputs

​Outputs

Inputs

Outputs