Picocolors vs Chalk: 1.0 KB vs 3.4 KB Gzip

I bundled the color APIs on this box: picocolors 1.1.1 came in at 1,035 B gzip against chalk 6.0.1 at 3,429 B. On a 200,000-call styled-string loop with FORCE_COLOR=1, picocolors ran about 9.4M ops/s and chalk about 4.2M.

Short answer: pick picocolors when you want the smallest ANSI dependency for a CLI or Vite plugin log line. Pick chalk when you already use nested chalk chains and want the wider chalk ecosystem. No affiliate links in this post.

What I tested

• Machine: 8 vCPU Intel Xeon / 15 GB RAM shared Linux cloud box, tested 6 Oct 2026 about 01:01–01:02 IST.
• Versions: picocolors 1.1.1, chalk 6.0.1, Node 22.20.0, esbuild 0.28.2 minify + gzip -9.
• Fixture: /workspace/bench-picocolors-chalk/. Size entries call red(bold), green, yellow, and cyan once each. Speed: same four styles concatenated, 200,000 iters × 9 rounds (median) after warm, with FORCE_COLOR=1 so both emit ANSI.
• Findings: node gzip 1,035 B vs 3,429 B; browser build 654 B vs 2,645 B. Median speed 21.274 ms vs 47.593 ms for 200k calls (about 9.40M vs 4.20M ops/s). Disk trees about 32 KB vs 108 KB.
• One surprise: with colors forced on, chalk was about 2.2× slower on this simple nest, not only larger. The gap is size-first, but the loop is not free either.
• Not tested: chalk nested template literals, chalk Level 3 truecolor, Windows terminals, browser CSS color helpers, TypeScript types surface, or chalk's full chain API beyond four styles.

How wide was the size gap?

Build (esbuild minify)	Minified	gzip -9
picocolors (node)	2,664 B	1,035 B
chalk 6 (node)	8,077 B	3,429 B
picocolors (browser)	1,286 B	654 B
chalk 6 (browser)	6,345 B	2,645 B

picocolors was about 3.3× smaller gzip on the node entry and about 4.0× smaller on the browser entry. Track shipped weight with size-limit vs bundlesize. After a swap, prune unused imports with knip vs depcheck.

How fast were the styled strings?

Path (200,000 calls, median of 9, FORCE_COLOR=1)	picocolors	chalk
red(bold)+green+yellow+cyan	21.274 ms (9.40M/s)	47.593 ms (4.20M/s)

On this box picocolors was roughly 2.2× faster on the nested style path. Formatters sit next to my dprint vs Prettier notes when I care about CLI output tooling, not page CSS.

Which one should you pick?

Situation	My pick
Vite / Rollup plugin that logs a few lines	picocolors
New CLI where every KB of deps matters	picocolors
App already deep in chalk chains	chalk
You need chalk's Level API / truecolor helpers	chalk
Bundle budget under about 1.5 KB gzip for colors	picocolors

Bottom line: picocolors 1.1.1 shipped 1.0 KB gzip against chalk 6.0.1 at 3.4 KB, and ran about 9.4M ops/s against 4.2M on a 200k styled-string fixture. Who should not switch: teams with a large chalk-colored codebase — rewrite cost beats a 2.4 KB gzip win. Lint the swap with Biome vs ESLint/Prettier. Test runner: Jest vs Vitest.

How this was made: I installed picocolors and chalk on the ShopperCove box, measured esbuild minify+gzip sizes (node and browser), timed FORCE_COLOR styled-string loops, and recorded the JSON. The write-up was drafted with AI help and checked against that output.

Sources

• https://github.com/alexeyraspopov/picocolors
• https://www.npmjs.com/package/picocolors
• https://github.com/chalk/chalk
• https://www.npmjs.com/package/chalk

Related

• https://www.shoppercove.com/blog/size-limit-vs-bundlesize
• https://www.shoppercove.com/blog/knip-vs-depcheck
• https://www.shoppercove.com/blog/dprint-vs-prettier
• https://www.shoppercove.com/blog/jest-vs-vitest
• https://www.shoppercove.com/blog/biome-vs-eslint-prettier
• https://www.shoppercove.com/blog/oxlint-vs-eslint
• https://www.shoppercove.com/blog/lefthook-vs-husky
• https://www.shoppercove.com/blog/pnpm-vs-npm-vs-bun