Math Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
81.94Accuracy
27
Apr 1, 2026
82.92Accuracy
27
Apr 1, 2026
91.25Accuracy
27
Apr 1, 2026
1,062Average@16
26
Apr 2, 2026
17.1Average@16
26
Apr 2, 2026
85.6Avg Pass@8
26
May 22, 2026
79.6Avg Pass@8
26
May 22, 2026
90.5Accuracy
26
Feb 26, 2026
70.42Accuracy
25
May 20, 2026
97.22Relative Overall Score
25
Jul 3, 2026
78.67Accuracy
25
Jun 30, 2026
58.86Accuracy
25
Feb 26, 2026
96.67Accuracy
25
Feb 26, 2026
35.6Mean Score (AIME 2025)
24
May 28, 2026
42Mean@16
24
May 28, 2026
43.8Mean@16
24
May 28, 2026
91.3Mean@16 Accuracy
24
May 28, 2026
88.6Pass@1
24
Jun 30, 2026
45.8Overall Accuracy
24
May 12, 2026
85Accuracy
24
May 12, 2026
20Accuracy (%)
24
May 12, 2026
23.3Accuracy
24
May 12, 2026
46.5Accuracy
24
Apr 21, 2026
54.48Accuracy
24
Apr 21, 2026
68.55Accuracy
24
Apr 21, 2026