List your company

Mathematical Reasoning

Benchmarks

DatasetSOTA methodMetricTrendResultsLast updated
93.3Pass@1 Accuracy
Mathematical Reasoning on AIME 2025 (Pass@1, #Tokens) performance trend
79
Jun 12, 2026
80.9Accuracy
Mathematical Reasoning on MATH (test) (Reward, Acc (%)) performance trend
73
Jun 24, 2026