Claude experiment day5
#869Comparing
claude-experiment-day5
(998493c
) withclaude-experiment-day4
(9522ef5
)-6%
- Improvement
- 6
- Regression
- 2
- Untouched
- 38
- Skipped
- 190
No successful run was found on claude-experiment-day4 ( 0945079) during the generation of this report, so ff305c1 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report.
Benchmarks
236 total
tests[proof_testing_typecheck]
benches/ci_benchmarking.rs-6%
134.3 ms142.8 ms
rust_rule_insert_loop[ops1000_funcs200]
benches/rust_api_benchmarking.rs-5%
334.6 µs353.7 µs
tests[hardboiled_conv1d_128]
benches/ci_benchmarking.rs+13%
3 s2.7 s
tests[hardboiled_conv1d_32]
benches/ci_benchmarking.rs+10%
975.6 ms883.4 ms
tests[hardboiled_conv1d_128]
benches/ci_benchmarking.rs+10%
2.5 s2.2 s
tests[hardboiled_conv1d_32]
benches/ci_benchmarking.rs+8%
876.6 ms809.5 ms
tests[cykjson]
benches/ci_benchmarking.rs+7%
440.4 ms412.8 ms
tests[proof_testing_unify]
benches/ci_benchmarking.rs+5%
18.7 ms17.8 ms
tests[eggcc-extraction]
benches/ci_benchmarking.rs+4%
1.5 s1.5 s
tests[eggcc-extraction]
benches/ci_benchmarking.rs+4%
1.5 s1.4 s
tests[herbie]
benches/ci_benchmarking.rs+4%
192.5 ms185.7 ms
tests[cykjson]
benches/ci_benchmarking.rs+4%
315.4 ms304.6 ms
tests[herbie]
benches/ci_benchmarking.rs+3%
167.1 ms161.8 ms
tests[python_array_optimize]
benches/ci_benchmarking.rs+3%
1.4 s1.4 s
tests[math-microbenchmark]
benches/ci_benchmarking.rs+2%
8.8 s8.6 s
tests[herbie-tutorial]
benches/ci_benchmarking.rs+2%
10.8 ms10.5 ms
tests[math-microbenchmark]
benches/ci_benchmarking.rs+2%
8 s7.9 s
tests[typeinfer]
benches/ci_benchmarking.rs+1%
107.3 ms105.8 ms
tests[stresstest_large_expr]
benches/ci_benchmarking.rs+1%
1.7 s1.7 s
tests[rectangle]
benches/ci_benchmarking.rs+1%
11 s10.9 s
tests[taylor51]
benches/ci_benchmarking.rs+1%
4.4 s4.4 s
rust_rule_tableaction_hot_path[facts50000_funcs200]
benches/rust_api_benchmarking.rs+1%
34.1 ms33.9 ms
rust_rule_insert_loop[ops1000_funcs200]
benches/rust_api_benchmarking.rs+1%
538.8 µs536 µs
tests[luminal-llama]
benches/ci_benchmarking.rs0%
637.6 ms634.6 ms
tests[luminal-llama]
benches/ci_benchmarking.rs0%
531.9 ms529.3 ms
Commits
Click on a commit to change the comparison rangeBase
claude-experiment-day4
9522ef5
-5.97%
Add Q15 journal entry (REVERTED): planner argmin monotonicity false syllogism
998493c
8 days ago
by yihozhang