Details for Benchmarks at
YJIT metrics from the yjit-bench suite using Ruby 634e0ac140.
Overall YJIT is 63.2% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 68.1% faster than CRuby!
On Railsbench specifically, YJIT is 68.1% faster than CRuby!
¶Performance on Headline Benchmarks
¶Memory Usage on Headline Benchmarks
¶Performance on Other Benchmarks
¶Memory Usage on Other Benchmarks

¶Performance on MicroBenchmarks
¶Memory Usage on MicroBenchmarks
¶Want Raw Graphs and CSV?
- Headlining benchmarks (x86_64)
- All benchmarks (x86_64)
- CSV speed results (x86_64)
- Headlining benchmarks (aarch64)
- All benchmarks (aarch64)
- CSV speed results (aarch64)
¶Benchmarks Speed Details

¶Benchmark Memory Usage Details
bench | CRuby 3.3.0dev mem (MiB) | YJIT 3.3.0dev mem (MiB) | Inline Code | Outlined Code | YJIT Mem overhead |
---|---|---|---|---|---|
activerecord | 48 | 54 | 1 | 1 | 11.5% |
chunky-png | 59 | 52 | 1 | 1 | -12.2% |
erubi-rails | 88 | 97 | 2 | 2 | 10.6% |
hexapdf | 151 | 289 | 2 | 2 | 91.8% |
liquid-c | 36 | 41 | 1 | 1 | 11.6% |
liquid-compile | 33 | 38 | 1 | 1 | 15.1% |
liquid-render | 34 | 38 | 1 | 1 | 12.8% |
lobsters | 275 | 330 | 8 | 7 | 19.9% |
50 | 56 | 1 | 1 | 10.8% | |
psych-load | 34 | 37 | 1 | 1 | 7.6% |
railsbench | 89 | 105 | 3 | 3 | 17.1% |
ruby-lsp | 94 | 132 | 5 | 5 | 39.1% |
sequel | 37 | 40 | 1 | 1 | 9.6% |
binarytrees | 25 | 26 | 1 | 1 | 4.5% |
erubi | 29 | 31 | 1 | 1 | 7.4% |
etanni | 23 | 23 | 1 | 1 | 2.2% |
fannkuchredux | 20 | 20 | 1 | 1 | 1.9% |
fluentd | 606 | 657 | 1 | 1 | 8.4% |
graphql | 39 | 42 | 1 | 1 | 7.9% |
graphql-native | 35 | 39 | 1 | 1 | 11.9% |
lee | 34 | 36 | 1 | 1 | 7.2% |
nbody | 20 | 20 | 1 | 1 | 1.7% |
optcarrot | 60 | 68 | 1 | 1 | 12.2% |
rack | 29 | 31 | 1 | 1 | 8.2% |
ruby-json | 21 | 21 | 1 | 1 | 2.2% |
rubykon | 47 | 48 | 1 | 1 | 3.3% |
tinygql | 27 | 30 | 1 | 1 | 8.1% |
30k_ifelse | 63 | 93 | 5 | 5 | 46.1% |
30k_methods | 51 | 59 | 2 | 2 | 15.3% |
cfunc_itself | 20 | 20 | 1 | 1 | 1.6% |
fib | 20 | 20 | 1 | 1 | 1.6% |
getivar | 20 | 20 | 1 | 1 | 1.6% |
keyword_args | 20 | 20 | 1 | 1 | 1.6% |
respond_to | 20 | 20 | 1 | 1 | 1.6% |
setivar | 20 | 20 | 1 | 1 | 1.6% |
setivar_object | 20 | 20 | 1 | 1 | 1.6% |
setivar_young | 20 | 20 | 1 | 1 | 1.7% |
str_concat | 46 | 61 | 1 | 1 | 33.5% |
throw | 20 | 20 | 1 | 1 | 1.6% |
Memory is shown in mebibytes (1024 * 1024 bytes.)
Older YJIT allocated an additional 256MiB for generated code. Current YJIT allocates executable memory on demand, so this overhead should no longer be present.
¶Number of Iterations and Warmups Tested
¶Benchmark YJIT Stats
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.
- aarch64_prod_ruby_no_jit
- aarch64_prod_ruby_with_yjit
- aarch64_yjit_stats
- x86_64_prod_ruby_no_jit
- x86_64_prod_ruby_with_yjit
- x86_64_yjit_stats