Open data for better minification decisions:

Community benchmarks

The Minifier
Benchmarks Project.

Regularly updated data for web page minifier comparison, testing effectiveness and efficiency on real-world websites.

–

Minifiers benchmarked

–

Websites tested

2

Minification modes

–

Last benchmark run

01 / The benchmarks

Less output, less time. See how web page minifiers compare.

Repository data

Loading results…

02 / Numbers with context

Read the notes

Real sites, not synthetic tests.

Dozens of websites, from the BBC to MDN and the ECMAScript specification. One general configuration per minifier, applied to the same source pages.

Warmed up, measured as fairly as possible.

Five timed runs per site after warm-up, using the median. Local minifiers run in their own processes and take turns in rotating order; times cover the minification call only.

Transparent about the trade-offs.

Failed sites count as unminified in size averages. Maximum minification is not a like-for-like comparison. Correctness caveats are documented.

03 / Built in the open

More data, better minification

A community-maintained collection of benchmarks, stewarded by Jens Oliver Meiert (context). Compare the tools, inspect the tests, and help make the next benchmark better. Minifier authors are invited to add their tools and become co-owners.

Contribute on GitHub