StableCascade_SuperResolutionControlnet - ComfyUI Built-in Node Documentation

This node is part of the experimental Stable Cascade group.

This node is part of the experimental Stable Cascade group. It prepares inputs for Stable Cascade super-resolution processing by encoding an input image with a VAE to create a controlnet input, and by generating empty (zero-filled) latent placeholders for stage C and stage B of the Stable Cascade pipeline.

Inputs

Parameter	Description	Data Type	Required	Range
image	The input image to be processed for super-resolution. Only the first 3 color channels (RGB) of the image are used for encoding.	IMAGE	Yes	-
vae	The VAE model used to encode the input image	VAE	Yes	-

Outputs

Output Name	Description	Data Type
controlnet_input	The VAE-encoded image representation suitable for controlnet input	IMAGE
stage_c	Placeholder (zero-filled) latent representation for stage C of Stable Cascade processing, with 16 channels and dimensions based on the input image size divided by 16	LATENT
stage_b	Placeholder (zero-filled) latent representation for stage B of Stable Cascade processing, with 4 channels and dimensions based on the input image size divided by 2	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): d9eff373ac7736f2e2f9788d1b43c04bb3212422aa1703d1d58ac512ce476925