Skip to content

Benchmark

FrontierCode 2.0

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

Models measured
5
Scores recorded
5
Sources
1
IQ weight
1.97

Score distribution

34.4%
Mean
43.8%
Median
14.1%
Std dev
15.1%
10th percentile
49.1%
90th percentile
15.1%
Min
15.1%49.1%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.97 for chat models, measured over 5 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Fable 5.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim49.1%
2GPT-6.2 Astra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim44.2%
3GPT-6.2 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim43.8%
4Mistral Large 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim19.7%
5Inkling 2 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim15.1%

Data credits: Data by LLM Stats