Claude experiment day2
#855Comparing
claude-experiment-day2
(7388953
) withclaude-experiment
(44f3e9e
)+46%
- Improvement
- 13
- Untouched
- 33
- Skipped
- 190
Different runtime environments detected
Some benchmarks with significant performance changes were compared across different runtime environments, which may affect the accuracy of the results. Learn more about regression causes.Benchmarks
236 total
tests[python_array_optimize]
benches/ci_benchmarking.rs+46%
2.2 s1.5 s
tests[python_array_optimize]
benches/ci_benchmarking.rs+41%
2.2 s1.6 s
rust_rule_match_overhead[rule_run_1]
benches/rust_api_benchmarking.rs+22%
9.5 ms7.7 ms
rust_rule_match_overhead[rule_run_1]
benches/rust_api_benchmarking.rs+22%
14.9 ms12.3 ms
rust_rule_fib[rule_run_1000]
benches/rust_api_benchmarking.rs+20%
337.8 ms280.5 ms
tests[hardboiled_conv1d_128]
benches/ci_benchmarking.rs+18%
3.8 s3.3 s
rust_rule_fib[rule_run_1000]
benches/rust_api_benchmarking.rs+17%
244.2 ms208.5 ms
tests[hardboiled_conv1d_128]
benches/ci_benchmarking.rs+17%
3.2 s2.7 s
tests[cykjson]
benches/ci_benchmarking.rs+13%
499.7 ms443.3 ms
tests[cykjson]
benches/ci_benchmarking.rs+12%
354.2 ms317.1 ms
rust_rule_tableaction_hot_path[facts50000_funcs200]
benches/rust_api_benchmarking.rs+9%
37.2 ms34.1 ms
rust_rule_tableaction_hot_path[facts50000_funcs200]
benches/rust_api_benchmarking.rs+9%
30 ms27.5 ms
tests[proof_testing_eqsat-basic]
benches/ci_benchmarking.rs+6%
36.6 ms34.5 ms
tests[hardboiled_conv1d_32]
benches/ci_benchmarking.rs+4%
1.1 s1.1 s
tests[proof_testing_unify]
benches/ci_benchmarking.rs+4%
22.3 ms21.4 ms
rust_rule_insert_loop[ops1000_funcs200]
benches/rust_api_benchmarking.rs+4%
346.8 µs333.4 µs
tests[herbie]
benches/ci_benchmarking.rs+3%
196.4 ms190.1 ms
tests[typeinfer]
benches/ci_benchmarking.rs+3%
131.9 ms127.9 ms
tests[luminal-llama]
benches/ci_benchmarking.rs+3%
671.3 ms651 ms
tests[hardboiled_conv1d_32]
benches/ci_benchmarking.rs+3%
999.3 ms973.6 ms
tests[extract-vec-bench]
benches/ci_benchmarking.rs+2%
80.5 ms78.7 ms
tests[luminal-llama]
benches/ci_benchmarking.rs+2%
545.3 ms536.1 ms
tests[repro-665-set-union]
benches/ci_benchmarking.rs+1%
960.8 ms950.2 ms
rust_rule_match_with_serialize[rule_run_1]
benches/rust_api_benchmarking.rs+1%
278.2 ms275.5 ms
tests[herbie]
benches/ci_benchmarking.rs+1%
172.5 ms171 ms
Commits
Click on a commit to change the comparison rangeBase
claude-experiment
44f3e9e
+45.96%
nits
7388953
16 hours ago
by yihozhang