Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
99.86Accuracy
97
Jun 9, 2026
80.78Pass@1
96
Feb 26, 2026
49.5Accuracy
94
May 1, 2026
57.29Pass@1
90
Jun 9, 2026
73.2Accuracy
90
Mar 4, 2026
83.9Accuracy
89
Jul 7, 2026
95.1Accuracy
89
May 11, 2026
99.2Accuracy
89
Jun 16, 2026
82.6Accuracy
89
Jul 7, 2026
81.2Accuracy
88
Jul 1, 2026
57.4Accuracy (avg@1)
87
Jun 2, 2026
96.5Accuracy
87
Feb 27, 2026
43.33Pass@1
86
Jun 9, 2026
50.8Accuracy @16
86
Jun 9, 2026
83.8Pass@1
86
Feb 26, 2026
83.75Pass@1 Accuracy
84
Jun 1, 2026
85Accuracy
84
Jun 1, 2026
93.3AIME 24 Accuracy
84
Feb 26, 2026
98.6Performance (%)
83
Jul 3, 2026
95Accuracy
83
Jun 9, 2026
61.8Accuracy
82
Jun 5, 2026
96.2Accuracy
82
Mar 10, 2026
82.2Accuracy
82
Jun 23, 2026
68.3Mean Score (k=8)
81
May 29, 2026
95Accuracy
81
Mar 24, 2026