EmptyHunyuanImageLatent - ComfyUI Built-in Node Documentation

The EmptyHunyuanImageLatent node creates a blank, all-zero latent space for Hunyuan image generation models.

The EmptyHunyuanImageLatent node creates a blank, all-zero latent space for Hunyuan image generation models. It produces an empty starting latent using the given width, height, and batch size, which can then be passed to downstream nodes in the workflow. The latent tensor contains 64 channels, and each spatial dimension equals the corresponding pixel dimension divided by 32.

Inputs

Parameter	Description	Data Type	Required	Range
width	The width of the generated latent image in pixels (default: 2048, step: 32)	INT	Yes	64 to MAX_RESOLUTION
height	The height of the generated latent image in pixels (default: 2048, step: 32)	INT	Yes	64 to MAX_RESOLUTION
batch_size	The number of latent samples to generate in a batch (default: 1)	INT	Yes	1 to 4096

Outputs

Output Name	Description	Data Type
LATENT	An empty latent tensor with 64 channels and dimensions of height ÷ 32 by width ÷ 32, ready for Hunyuan image processing	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): 31fc10d43c224810709870cf40256b6fccd4743445ea9d98d148d443bc591d7a