Skip to content

Benchmark

FrontierCode

Coding benchmarks compare models on programming tasks. Scores are percentages.

Models measured
4
Scores recorded
4
Sources
1
IQ weight
No weight published

Score distribution

50%
Mean
53.4%
Median
8.3%
Std dev
38.8%
10th percentile
61.3%
90th percentile
38.8%
Min
38.8%61.3%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Grok 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim61.3%
2Claude Opus 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim53.4%
3Claude Fable 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim46.3%
4Claude Sonnet 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim38.8%

Data credits: Data by LLM Stats