Trie of 50k words

Insert 50k dictionary words into a trie; perform 10k prefix queries.

Runtime · median per inner-loop window

median of 10 runs

Native Gocompiled

8.60 ms0.02×

Piko interpbytecode VM

527 msbaseline

CPython 3.13bytecode VM

547 ms1.04×

PyPy 7.3tracing JIT

52.9 ms0.10×

Ttengobytecode VM

1.47 s2.79×

Sscriggobytecode VM

unsupported in scriggo

n/a

Mmvmbytecode VM

1.28 s2.43×

YyaegiAST walker

1.33 s2.51×

Runner	N	Compile	Runtime	P95	Stddev	RSS	vs piko	Status
Native Gocompiled	10	183 ms	8.60 ms	8.78 ms	72.8 µs	68 MiB	0.02×	OK
Piko interpbytecode VM	10	1.21 ms	527 ms	531 ms	2.36 ms	142 MiB	1.00×	OK
CPython 3.13bytecode VM	10	381 µs	547 ms	570 ms	9.25 ms	n/a	1.04×	OK
PyPy 7.3tracing JIT	10	318 µs	52.9 ms	54.0 ms	663 µs	n/a	0.10×	OK
tengobytecode VM	10	270 µs	1.47 s	1.52 s	21.3 ms	2.22 GiB	2.79×	OK
scriggobytecode VM	0	n/a	n/a	n/a	n/a	n/a	n/a	unsupported
mvmbytecode VM	10	406 µs	1.28 s	1.31 s	11.2 ms	67 MiB	2.43×	OK
yaegiAST walker	10	470 µs	1.33 s	1.44 s	61.0 ms	66 MiB	2.51×	OK

Workload & symmetry rules

Workload

Build a 26-way trie from 50,000 lowercase ASCII words. Run 10,000 random prefix-existence queries. Print the count of hits.

Source code