EmptySD3LatentImage - ComfyUI Built-in Node Documentation

Complete documentation for the EmptySD3LatentImage node in ComfyUI. Learn its inputs, outputs, parameters and usage.

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

The EmptySD3LatentImage node creates a blank latent image tensor specifically formatted for Stable Diffusion 3 models. It generates a tensor filled with zeros that has the correct dimensions and structure expected by SD3 pipelines. This is commonly used as a starting point for image generation workflows.

Inputs

Parameter	Data Type	Required	Range	Description
width	INT	Yes	16 to MAX_RESOLUTION (step: 16)	The width of the output latent image in pixels (default: 1024)
height	INT	Yes	16 to MAX_RESOLUTION (step: 16)	The height of the output latent image in pixels (default: 1024)
batch_size	INT	Yes	1 to 4096	The number of latent images to generate in a batch (default: 1)

Outputs

Output Name	Data Type	Description
LATENT	LATENT	A latent tensor containing blank samples with SD3-compatible dimensions