Rust Parsing Benchmarks

This repo tries to assess Rust parsing performance.

crate	parser type	action code	integration	input type	precedence climbing	parameterized rules	streaming input
chumsky	combinators	in source	library	&str	?	?	?
combine	combinators	in source	library	&str	?	?	?
grmtools	CFG	in grammar	library	?	?	?	?
lalrpop	LR(1)	in grammar	build script	&str	No	Yes	No
lelwel	LL(1)	in grammar	build script	&str	Yes	Yes	No
logos	lexer	in source	proc macro	&str, &[u8]	?	?	?
nom	combinators	in source	library	&str, &[u8], custom	No	Yes	Yes
parol	LL(k)/LALR(R)	in grammar	build script	&str	No	?	No
peg	PEG	in grammar	proc macro (block)	&str, &[T], custom	Yes	Yes	No
pest	PEG	external	proc macro (file)	&str	Yes	No	No
winnow	combinators	in source	library	&str, &[T], custom	No	Yes	Yes
yap	combinators	in source	library	&str, &[T], custom	No	Yes	?

Formerly, we compared:

• pom: lack of notoriety

Results

Name	Overhead (release)	Build (debug)	Parse (release)	Downloads	Version
null	0 KiB	196ms	4ms	-	-
grmtools	2,573 KiB	12s	167ms		v0.13.10
chumsky	297 KiB	4s	85ms		v0.10.0
combine	184 KiB	4s	47ms		v3.8.1
lalrpop	1,526 KiB	12s	37ms		v0.22.1
lelwel	152 KiB	4s	10ms		v0.7.1
logos	90 KiB	5s	20ms		v0.15.0
nom	98 KiB	3s	59ms		v8.0.0
parol	1,718 KiB	14s	262ms		v3.0.1
peg	85 KiB	2s	21ms		v0.8.5
pest	128 KiB	4s	64ms		v2.8.0
serde_json	55 KiB	3s	14ms		v1.0.140
winnow	79 KiB	2s	27ms		v0.7.4
yap	65 KiB	447ms	31ms		v0.12.0

System: Linux 5.4.0-170-generic (x86_64), rustc 1.85.1 (4eb161250 2025-03-15) w/ -j 8

Note:

• For more "Parse (release)" comparisons, see parser_benchmarks
• Parsers have not been validated and might have differing levels of quality (#5)

Running the Benchmarks

$ ./bench.py
$ ./format.py