Math Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
3.8Speedup
32
Jul 1, 2026
30Accuracy (%)
32
Jun 16, 2026
82.5Accuracy (%)
32
Jun 16, 2026
50.6Score
32
Mar 25, 2026
75.96Accuracy
31
Jun 15, 2026
74.4Pass@1
30
Jun 30, 2026
55.94Accuracy
30
Apr 10, 2026
68.6Pass@1
30
Mar 16, 2026
78.8Score
30
May 19, 2026
98.4During-task Accuracy
29
Feb 26, 2026
99.3Accuracy
29
Jun 30, 2026
87.5Avg@16
29
May 28, 2026
69.7Mean@12 Score
28
Jun 11, 2026
77.5Mean@12 Score
28
Jun 11, 2026
92.7Mean Score @12
28
Jun 11, 2026
42.48EM Score
28
May 19, 2026
94.6AIME25 Accuracy
28
Apr 24, 2026
100Final Accuracy
28
Feb 26, 2026
47.29Avg@8 Accuracy
27
Apr 29, 2026
11.25Accuracy (avg@16)
27
Apr 29, 2026
18.33Accuracy (avg@16)
27
Apr 29, 2026
71.08Pass@8
27
Apr 29, 2026
87.8Success Rate (pass@4)
27
Apr 29, 2026
36.67Pass@16
27
Apr 29, 2026
36.67Pass@16
27
Apr 29, 2026