Word frequency

Tokenise a 1 MiB ASCII corpus and count word occurrences into a hash map.

Compile time · median (cold)

median of 10 runs

Native Gocompiled

181 ms104×

Piko interpbytecode VM

1.74 msbaseline

CPython 3.13bytecode VM

452 µs0.26×

PyPy 7.3tracing JIT

387 µs0.22×

Ttengobytecode VM

332 µs0.19×

Sscriggobytecode VM

653 µs0.37×

Mmvmbytecode VM

506 µs0.29×

YyaegiAST walker

485 µs0.28×

Runner	N	Compile	Runtime	P95	Stddev	RSS	vs piko	Status
Native Gocompiled	10	181 ms	18.0 ms	18.3 ms	154 µs	68 MiB	104×	OK
Piko interpbytecode VM	10	1.74 ms	182 ms	185 ms	1.85 ms	307 MiB	1.00×	OK
CPython 3.13bytecode VM	10	452 µs	313 ms	324 ms	6.94 ms	n/a	0.26×	OK
PyPy 7.3tracing JIT	10	387 µs	40.8 ms	46.0 ms	1.89 ms	n/a	0.22×	OK
tengobytecode VM	10	332 µs	788 ms	936 ms	61.5 ms	5.36 GiB	0.19×	OK
scriggobytecode VM	10	653 µs	417 ms	432 ms	7.69 ms	1.12 GiB	0.37×	OK
mvmbytecode VM	10	506 µs	696 ms	755 ms	31.4 ms	71 MiB	0.29×	OK
yaegiAST walker	10	485 µs	463 ms	500 ms	14.3 ms	72 MiB	0.28×	OK

Workload & symmetry rules

Workload

Read a 1 MiB lorem corpus. Tokenise on whitespace + ASCII punctuation. Count occurrences in a map. Print the top-10 by frequency.

Source code