YJIT Benchmarks

Details for Benchmarks at 2022-12-05 20:10:10

Overall YJIT is 36.9% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 33.5% faster than CRuby!
The basic "faster" measurement is the geomean of all "headlining" x86 benchmarks on this page.

Headlining Benchmarks

These are "headlining" because the "overall" speedup above is based on these benchmarks specifically.

Speed of each Ruby implementation (iterations/second) relative to the CRuby interpreter. Higher is better.

Other Benchmarks

Speed of each Ruby implementation (iterations/second) relative to the CRuby interpreter. Higher is better.

MicroBenchmarks

Speed of each Ruby implementation (iterations/second) relative to the CRuby interpreter. Higher is better.

Want Raw Graphs and CSV?

Benchmarks Speed Details

Benchmark Memory Usage Details

Number of Iterations and Warmups Tested

bench No JIT warmups No JIT iters YJIT warmups YJIT iters
activerecord 5 193 5 193
hexapdf 5 15 5 15
liquid-render 5 156 5 156
mail 5 139 5 139
psych-load 5 15 5 15
railsbench 5 15 5 15
ruby-lsp 5 277 5 277
binarytrees 5 84 5 84
chunky_png 5 32 5 32
erubi 5 69 5 69
erubi_rails 5 1133 5 1133
etanni 5 42 5 42
fannkuchredux 5 21 5 21
lee 5 22 5 22
nbody 5 270 5 270
optcarrot 5 15 5 15
rubykon 5 15 5 15
30k_ifelse 5 49 5 49
30k_methods 5 20 5 20
cfunc_itself 5 489 5 489
fib 5 327 5 327
getivar 5 435 5 435
keyword_args 5 383 5 383
respond_to 5 697 5 697
setivar 5 1360 5 1360
str_concat 5 497 5 497

Different Ruby configurations want different amounts of warmup. With no JIT, CRuby needs hardly any. YJIT and MJIT 3.0 both warm up quite quickly, while MJIT in 3.1 often slows down for a time as it compiles, after an unpredictable delay.

Benchmark YJIT Stats

Hover your cursor over the benchmark names for descriptions of each benchmark.

bench Exit Report Inline Outlined Comp iSeqs Comp Blocks Inval Inval Ratio Bind Alloc Bind Set Const Bumps
activerecord (click) 2271706 2270583 25 359 6 1% 0 0 0
hexapdf (click) 1707265 1706140 820 10222 216 2% 0 0 0
liquid-render (click) 576783 575568 156 1847 33 1% 0 0 0
mail (click) 1249688 1248201 373 4646 176 3% 0 0 0
psych-load (click) 784921 783116 70 649 19 2% 0 0 0
railsbench (click) 5507864 5506130 1779 14261 482 3% 0 0 0
ruby-lsp (click) 5350927 5349558 6428 41473 1522 3% 14154 0 0
binarytrees (click) 131047 131044 16 140 4 2% 0 0 0
chunky_png (click) 799722 798290 89 1175 31 2% 0 0 0
erubi (click) 731340 730631 13 147 4 2% 0 0 0
erubi_rails (click) 4648346 4646600 329 2800 111 3% 0 0 0
etanni (click) 131842 131421 13 108 4 3% 0 0 0
fannkuchredux (click) 143253 142154 14 273 4 1% 0 0 0
lee (click) 803263 802911 79 812 25 3% 0 0 0
nbody (click) 143845 142438 15 245 6 2% 0 0 0
optcarrot (click) 563064 561272 203 4439 93 2% 0 0 0
rubykon (click) 271645 271293 146 1641 31 1% 0 0 0
30k_ifelse (click) 5280092 5280038 9268 57200 4 0% 0 0 0
30k_methods (click) 2028160 2026418 5787 19427 4 0% 0 0 0
cfunc_itself (click) 128985 127266 13 123 4 3% 0 0 0
fib (click) 127556 126066 13 112 4 3% 0 0 0
getivar (click) 129334 127585 13 139 4 2% 0 0 0
keyword_args (click) 129763 127967 14 125 4 3% 0 0 0
respond_to (click) 122269 120110 13 139 5 3% 0 0 0
setivar (click) 128373 126769 13 119 4 3% 0 0 0
str_concat (click) 122024 119967 15 138 6 4% 0 0 0

YJIT stats correspond to the YJIT stats exit report.

Note: currently, all stats are collected on x86_64, not ARM.

Raw JSON data files

All graphs and table data in this page comes from processing these data files, which come from benchmark runs.