Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
29.9Accuracy
13
Mar 16, 2026
40Accuracy
13
Mar 6, 2026
50.98Best-of-64 AIME Accuracy
13
Mar 19, 2026
35Accuracy
13
Mar 4, 2026
48.8Accuracy
13
Mar 4, 2026
54.5Avg@16 Score
13
Mar 4, 2026
75.1Avg@16
13
Mar 4, 2026
25.3Avg@16
13
Mar 4, 2026
73.75Avg. Pass@1
13
Feb 26, 2026
44.01Average Score
13
Feb 26, 2026
81.1Accuracy (Original)
13
Feb 26, 2026
0.88Accuracy (Original)
13
Feb 26, 2026
54.3Accuracy (L=128)
13
May 15, 2026
83.8P@1
13
Feb 26, 2026
33.5P@1
13
Feb 26, 2026
15.2P@1
13
Feb 26, 2026
93Accuracy
13
Feb 26, 2026
97.8Accuracy
13
Feb 26, 2026
78.51Accuracy
13
Feb 26, 2026
2.66Speedup (x)
13
Feb 26, 2026
58.48Pass@1 Accuracy
12
Jul 9, 2026
31.25Pass@1 Accuracy
12
Jul 9, 2026
88.4Pass@1 Accuracy
12
Jul 9, 2026
87.5Pass@1 Accuracy
12
Jul 9, 2026
41.04Pass@1 Accuracy
12
Jul 9, 2026