StableCascade_EmptyLatentImage - ComfyUI Built-in Node Documentation

The StableCascade_EmptyLatentImage node creates empty latent tensors for Stable Cascade models. It generates two separate latent representations - one for stage C and another for stage B - with appropriate dimensions based on the input resolution and compression settings. This node provides the starting point for the Stable Cascade generation pipeline.

Inputs

Parameter	Description	Data Type	Required	Range
`width`	The width of the output image in pixels (default: 1024, step: 8)	INT	Yes	256 to MAX_RESOLUTION
`height`	The height of the output image in pixels (default: 1024, step: 8)	INT	Yes	256 to MAX_RESOLUTION
`compression`	The compression factor that determines the latent dimensions for stage C (default: 42, step: 1). This is an advanced parameter.	INT	Yes	4 to 128
`batch_size`	The number of latent samples to generate in a batch (default: 1)	INT	No	1 to 4096

Note: The compression value controls the stage C latent size: its height and width are the input height and width divided by compression. The stage B latent always uses a fixed compression of 4.

Outputs

Output Name	Description	Data Type
`stage_c`	The stage C latent tensor with dimensions [batch_size, 16, height//compression, width//compression]	LATENT
`stage_b`	The stage B latent tensor with dimensions [batch_size, 4, height//4, width//4]	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): f336f87d0ec14b3716efda2cfaa194b1f80707d64821bb56ade7d88d9bd5b53f

​Inputs

​Outputs

Inputs

Outputs