Skip to main content
The EmptyCosmosLatentVideo node creates an empty latent video tensor with specified dimensions. It generates a zero-filled latent representation that can be used as a starting point for video generation workflows, with configurable width, height, length, and batch size parameters.

Inputs

ParameterDescriptionData TypeRequiredRange
widthThe width of the latent video in pixels (default: 1280, must be divisible by 16)INTYes16 to MAX_RESOLUTION
heightThe height of the latent video in pixels (default: 704, must be divisible by 16)INTYes16 to MAX_RESOLUTION
lengthThe number of frames in the latent video (default: 121, must be divisible by 8)INTYes1 to MAX_RESOLUTION
batch_sizeThe number of latent videos to generate in a batch (default: 1)INTNo1 to 4096

Outputs

Output NameDescriptionData Type
samplesThe generated empty latent video tensor with zero valuesLATENT
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): ed7404f31b4b9fb2c098579f8e96e55befb3c8ed045d4175385d01a7b42035e2