Math
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
54.1Score
10
Feb 26, 2026
63.33Score
10
Feb 26, 2026
5.832PPL
10
Feb 26, 2026
38.33Accuracy (MMath)
9
Jun 26, 2026
0.489Accuracy (AIME 25)
9
Jun 2, 2026
89.08EM Score
9
Mar 20, 2026
15.59Score
9
Mar 4, 2026
5.27Accuracy
9
Mar 4, 2026
17.66Score
8
May 29, 2026
38.87Avg@4 Score
8
May 12, 2026
60.8Math Score
8
Feb 26, 2026
61.9Pass@1
7
Jul 2, 2026
90.5Pass@1 Score
7
Jul 2, 2026
70.5Pass@1
7
Jul 2, 2026
88.8Pass@1
7
Jul 2, 2026
89.8Pass@1
7
Jul 2, 2026
31.6Score
7
Jun 11, 2026
63.9Score
7
Jun 11, 2026
70.5Score
7
Jun 11, 2026
87.5Pass@1
7
Feb 26, 2026
91.4Pass@1
7
Feb 26, 2026
50.46Original Accuracy
6
Jun 19, 2026
55.1Original Accuracy
6
Jun 19, 2026
66.35Original Accuracy
6
Jun 19, 2026
62.35Original Accuracy
6
Jun 19, 2026