Khroma

Khroma provides an AI-powered color tool for designers to discover, search, and save color palettes. It learns which colors a user likes by building a neural-network powered model from chosen colors, then generates colors the user might prefer and blocks others. The system operates in the user’s browser, allowing immediate, local interaction without requiring external steps or downloads. The design focuses on providing a personalized palette experience directly in the browsing environment.

The tool trains on user input to tailor color outputs. By selecting a set of colors, users train a neural network powered algorithm to generate colors they like and block ones they don’t, right in their browser. It aims to create infinite color combinations based on a user’s preferences, derived from a broad set of palettes learned from the web.

It offers various views for palettes such as typography, gradient, palette, or a custom image. The cataloged results can be searched and explored to fit specific design needs, making it possible to navigate large sets of color options.

Users can save combinations to an unlimited collection, access color names, hex codes, RGB values, CSS code, and WCAG accessibility ratings for each pair, supporting reuse and accessibility considerations. The process centers on enabling designers to repeatedly generate and store preferred color combos for future reference.

Use cases

• Teams needing a searchable palette library with accessibility data