Repeat Into Grid (latent)

Documentation

• Class name: Repeat Into Grid (latent)
• Category: Bmad/latent
• Output node: False

This node tiles the input latent samples into a grid of configurable dimensions, effectively repeating the samples across a specified number of rows and columns to create a larger, grid-like structure.

Input types

Required

• samples
  • The latent samples to be tiled into a grid. This input is crucial for determining the pattern and content of the resulting grid.
  • Comfy dtype: LATENT
  • Python dtype: Dict[str, torch.Tensor]
• columns
  • Specifies the number of columns in the grid. This affects the horizontal repetition of the samples.
  • Comfy dtype: INT
  • Python dtype: int
• rows
  • Specifies the number of rows in the grid. This affects the vertical repetition of the samples.
  • Comfy dtype: INT
  • Python dtype: int

Output types

• latent
  • Comfy dtype: LATENT
  • The resulting latent samples arranged in a grid, with the original samples repeated according to the specified rows and columns.
  • Python dtype: Tuple[Dict[str, torch.Tensor]]

Usage tips

• Infra type: GPU
• Common nodes: unknown

Source code

class RepeatIntoGridLatent:
    """
    Tiles the input samples into a grid of configurable dimensions.
    """

    def __init__(self):
        pass

    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"samples": ("LATENT",),
                             "columns": grid_len_INPUT,
                             "rows": grid_len_INPUT,
                             }}

    RETURN_TYPES = ("LATENT",)
    FUNCTION = "repeat_into_grid"
    CATEGORY = "Bmad/latent"

    def repeat_into_grid(self, samples, columns, rows):
        s = samples.copy()
        samples = samples['samples']
        tiled_samples = samples.repeat(1, 1, rows, columns)
        s['samples'] = tiled_samples
        return (s,)