This ComfyUI workflow extracts a cohesive color palette from any input image and renders it as a clean, full-resolution color bar. It’s built around a simple graph: LoadImage ingests your source, a custom palette processor (node ID 0650f674-0daf-4978-ac64-0e41060db106) analyzes the image to select a set number of representative colors and their relative dominance, and SaveImage writes the resulting palette bar to disk. A MarkdownNote in the canvas documents the controls so you can quickly adjust behavior without guessing.

Technically, the custom node performs color reduction to N key colors and can size each strip either proportionally to its estimated dominance (Weighted) or evenly (Equal). When Weighted is ON, the node uses dominance percentages to assign band widths; when OFF, all colors receive the same width. This makes the workflow useful for design and art direction tasks: you can pull brand palettes, build mood references from photography or film frames, and export a neat, shareable palette image—without running a diffusion model or heavy GPU inference.