Skip to content

Benchmark

FrontierBench v0.1

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

Models measured
4
Scores recorded
6
Sources
2
IQ weight
No weight published

Score distribution

36.7%
Mean
43.3%
Median
8.2%
Std dev
33.8%
10th percentile
44.4%
90th percentile
21.1%
Min
21.1%44.4%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Claude Opus 5Anthropic self-run · xhigh· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.44.4%
2Claude Opus 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim43.3%
3Claude Opus 5Harbor-reported· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.43.3%
4GPT-5.6 SolCodex· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.34.4%
5Claude Fable 5Opus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.33.8%
6Claude Opus 4.8Opus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.21.1%

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