StableCascade_StageC_VAEEncode - ComfyUI Built-in Node Documentation

The StableCascade_StageC_VAEEncode node processes an input image through a VAE encoder to generate latent representations for the Stable Cascade model. It first resizes the image based on a compression factor and the VAE’s downscale ratio, then encodes the resized image. The node outputs two latent tensors: one for stage C (the actual encoded result) and one for stage B (a zero-filled placeholder).

Inputs

Parameter	Description	Data Type	Required	Range
`image`	The input image to be encoded into latent space	IMAGE	Yes	-
`vae`	The VAE model used for encoding the image	VAE	Yes	-
`compression`	The compression factor applied to the image before encoding. The image dimensions are divided by this value, then multiplied by the VAE’s downscale ratio. (default: 42)	INT	No	4-128

Outputs

Output Name	Description	Data Type
`stage_c`	The encoded latent representation for stage C of the Stable Cascade model	LATENT
`stage_b`	A placeholder latent representation for stage B. Currently returns a zero-filled tensor with dimensions calculated from the input image size.	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): 6ebdd035349744bc76ab294f16504672d3e79883fccc86d6ea057b4b15faef05

StableCascade_EmptyLatentImage - ComfyUI Built-in Node Documentation

EmptySD3LatentImage - ComfyUI Built-in Node Documentation

⌘I

​Inputs

​Outputs

Inputs

Outputs