Sharkbench

Benchmarking programming languages and web frameworks

Computation

This benchmark tests how fast a programming language can perform mathematical computations without any I/O or memory allocation. We are using the Leibniz formula to approximate the value of PI.

OS: LinuxCPU: Intel Core Ultra 5 325Last Updated: 2026-10-14

Language	Time How long does it take to approximate PI? (lower is better, log scale)	Memory How much memory (RAM) does the program use? (lower is better)	Build Time Fetching dependencies + compiling, excluding SDK setup (lower is better)
C (GCC)	1.00 s	2.2 MB	2.13 s
Julia	1.00 s	356.7 MB	1.26 s
Java (Temurin)	1.00 s	65.9 MB	14.84 s
Rust	1.00 s	2.9 MB	4.12 s
D	1.00 s	5.7 MB	4.18 s
Zig	1.00 s	1.1 MB	22.49 s
Kotlin (Native)	1.01 s	5.2 MB	67.97 s
Swift	1.02 s	10.8 MB	5.16 s
C# (.NET)	1.16 s	34.8 MB	4.49 s
Go	1.36 s	3.8 MB	6.18 s
Java (Semeru)	1.48 s	65.4 MB	15.74 s
Dart (AOT)	1.94 s	8.3 MB	4.94 s
Dart (JIT)	1.95 s	156.3 MB	1.72 s
Python (PyPy)	2.17 s	77.3 MB	1.25 s
JavaScript (Node.js)	3.24 s	66.4 MB	1.09 s
JavaScript (Bun)	3.86 s	45.3 MB	1.30 s
Elixir	18.82 s	66.4 MB	1.23 s
Ruby (JIT)	28.82 s	18.9 MB	1.12 s
Ruby	67.47 s	17.8 MB	1.29 s
Python	99.28 s	11.1 MB	1.28 s