Visit ComfyUI Online for ready-to-use ComfyUI environment
Generate custom latent tensors with specified width and height ratios for AI art creation, simplifying space creation process.
The Empty Latent Ratio Custom SDXL node is designed to generate latent tensors with custom dimensions based on specified width and height ratios. This node is particularly useful for AI artists who need to create latent spaces tailored to specific aspect ratios, ensuring that the generated images fit their desired dimensions. By leveraging predefined ratios or calculating new ones, this node provides flexibility and precision in generating latent tensors, which are essential for various AI art generation tasks. The node simplifies the process of creating latent spaces by automatically adjusting the dimensions based on the input ratios, making it easier for you to focus on the creative aspects of your work.
The width
parameter specifies the width of the latent space to be generated. It accepts integer values with a default of 1024, a minimum of 1, and a maximum of 8192. This parameter determines the horizontal dimension of the latent tensor and plays a crucial role in defining the aspect ratio of the generated latent space.
The height
parameter specifies the height of the latent space to be generated. It accepts integer values with a default of 1024, a minimum of 1, and a maximum of 8192. This parameter determines the vertical dimension of the latent tensor and, along with the width, defines the overall aspect ratio of the latent space.
The batch_size
parameter determines the number of latent tensors to be generated in a single batch. It accepts integer values with a default of 1, a minimum of 1, and a maximum of 64. This parameter is useful for generating multiple latent spaces simultaneously, which can be beneficial for batch processing or creating variations of the same latent space.
The LATENT
output parameter represents the generated latent tensor(s). This output is a dictionary containing a key samples
that holds the latent tensor(s) with the specified dimensions. The latent tensor is a multi-dimensional array filled with zeros, which serves as a blank canvas for further processing or image generation tasks. The dimensions of the tensor are determined by the input width, height, and batch size, ensuring that the output matches the desired aspect ratio and quantity.
© Copyright 2024 RunComfy. All Rights Reserved.