Skip to content

Benchmark

AIME

Math benchmarks measure how well models solve maths problems. Scores are percentages.

Models measured
2
Scores recorded
2
Sources
1
IQ weight
No weight published

Score distribution

47.4%
Mean
57.5%
Median
10.1%
Std dev
37.3%
10th percentile
57.5%
90th percentile
37.3%
Min
37.3%57.5%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Aihubmix Phi 4 Mini (reasoning)· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim57.5%
2MiMo V2.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim37.3%

Data credits: Data by LLM Stats