Khroma

Khroma's core strength is its personalization loop. Instead of serving you random combinations, it trains a neural network on the colors you pick during onboarding and uses that to generate palettes that match your taste. You can also explicitly block colors you dislike, which makes the results feel curated rather than algorithmic. The underlying model has learned from thousands of popular human-made palettes across the internet, so the output tends to feel more designerly than purely mathematical. The viewing options are practical: you can see any palette as typography, a gradient, a flat palette, or applied to a custom image. That last option is useful for judging how a color combination actually looks in context. The search tool is unusually granular — you can filter the generator by hue, tint, value, color, hex, and RGB values. If you need a palette in a specific direction, you can steer the generator instead of just hitting refresh. The saved library is unlimited and each palette includes color names, hex codes, RGB values, CSS code, and a WCAG accessibility rating for each color pair. That accessibility rating is the feature that elevates Khroma from a moodboarding toy to something you can use before shipping UI. You can check contrast ratios before committing a color combination to a production design. Where Khroma falls short is collaboration and integration. It's a single-user experience with no shared libraries, no real-time team features, and no documented Figma, Sketch, or Adobe plugin. Your only export is CSS, which is fine for web work but limiting for other workflows. There's also no API mentioned in the provided evidence, so developers can't programmatically generate palettes. And it's browser-only — no desktop app, no mobile app. For solo designers doing quick, personal color exploration with accessibility checks built in, Khroma is a practical free tool. For teams or production workflows that need shared assets or deep integration with design tools, it won't fit.