Math Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
61.8MATH500 Score
11
May 20, 2026
57.2Accuracy (Olympiad)
11
May 18, 2026
40.8Accuracy
11
May 18, 2026
45.4Score
11
May 12, 2026
8.15Score
11
May 12, 2026
74.8Accuracy
10
Jul 9, 2026
81.9Accuracy
10
Jul 3, 2026
79.5Accuracy
10
May 12, 2026
59.05Accuracy@1
10
Mar 4, 2026
36.76Acc@1
10
Mar 4, 2026
82.18Acc@16
10
Mar 4, 2026
40Accuracy @16
10
Mar 4, 2026
75.1MATH Score
10
Feb 26, 2026
1.26Accuracy
10
Feb 27, 2026
14.2Accuracy
10
Feb 27, 2026
80.1Avg@8 Score
10
May 22, 2026
0.831EM
10
Feb 26, 2026
41.5Pass@1
9
Jun 5, 2026
88Coverage
9
May 4, 2026
39.3Coverage
9
May 4, 2026
86Coverage
9
May 4, 2026
64.04Pass@k Score
9
Apr 29, 2026
100Score
9
May 28, 2026
0.92Task Accuracy
9
Feb 26, 2026
94.8Pass@16
9
Feb 26, 2026