Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
5.12Accepted Length (τ)
12
Jul 7, 2026
83.04Accuracy@1
12
Jul 7, 2026
79.8Accuracy
12
Jul 7, 2026
79.56AVG@32
12
Jul 7, 2026
77.1AIME24 Accuracy
12
Jul 3, 2026
79.4AIME24 Score
12
Jul 3, 2026
83.3Accuracy
12
Jul 1, 2026
86Accuracy
12
Jul 1, 2026
66.26GSM8K EM
12
Jul 3, 2026
7Samples Required
12
Jul 1, 2026
128Samples Required
12
Jul 1, 2026
53.2Accuracy
12
Jun 30, 2026
92.1Accuracy
12
Jun 30, 2026
36.67Pass@1
12
Jun 30, 2026
79.2Accuracy
12
Jun 30, 2026
92.65Accuracy
12
Jun 30, 2026
55.57Mean@16
12
Jun 23, 2026
18.33Mean@16
12
Jun 23, 2026
18.96Mean@16
12
Jun 23, 2026
67.56Mean@16
12
Jun 23, 2026
36.29Pass@1
12
Jun 23, 2026
96.84GSM8K Accuracy
12
Jun 23, 2026
34.18Avg@8 Accuracy
12
Jun 23, 2026
50.1Avg@8 Accuracy
12
Jun 23, 2026
71.25Avg@8 Accuracy
12
Jun 23, 2026