Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.comfy.org/llms.txt

Use this file to discover all available pages before exploring further.

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub
The HyperTile node applies a tiling technique to the attention mechanism in diffusion models to optimize memory usage during image generation. It divides the latent space into smaller tiles and processes them separately, then reassembles the results. This allows for working with larger image sizes without running out of memory.

Inputs

ParameterData TypeRequiredRangeDescription
modelMODELYes-The diffusion model to apply the HyperTile optimization to
tile_sizeINTNo1 - 2048The target tile size for processing (default: 256). The effective tile size is rounded down to a multiple of 8, with a minimum of 32.
swap_sizeINTNo1 - 128Controls how the tiles are rearranged during processing to improve efficiency (default: 2)
max_depthINTNo0 - 10The maximum depth level (resolution scale) to apply tiling. A value of 0 applies tiling only at the highest resolution (default: 0)
scale_depthBOOLEANNoTrue / FalseWhen enabled, the tile size is scaled proportionally at deeper depth levels. This can help maintain quality at lower resolutions (default: False)

Outputs

Output NameData TypeDescription
modelMODELThe modified model with HyperTile optimization applied

Source fingerprint (SHA-256): ea770eea3c628ffb166c2a15a6f75757cebd99c3fdd6225f041b3ef01a0e73c7