Skip to content

Benchmark

OpenRCA

Coding benchmarks compare models on programming tasks. Scores are percentages. Higher is better.

Models measured
2
Scores recorded
7
Sources
2
IQ weight
No weight published

Score distribution

32.5%
Mean
33.6%
Median
3.7%
Std dev
27.9%
10th percentile
34.9%
90th percentile
26.4%
Min
26.4%37.3%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Claude Opus 4.6Opus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted · Banking· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.37.3%
2Claude Opus 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim34.9%
3Claude Opus 4.6Opus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted · Overall· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.34.9%
4Claude Opus 4.6Opus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted · Market· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.33.6%
5Claude Opus 4.6Opus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted · Telecom· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.32.7%
6Claude Sonnet 4.6author harness · adaptive thinking · high effort · 3-run mean· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.27.9%
7Claude Sonnet 4.6author harness · adaptive thinking · max effort · 3-run mean· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.26.4%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats