Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
95.6Pass@32
7
May 13, 2026
43.3Pass@32
7
May 13, 2026
62.4Pass@32
7
May 13, 2026
63.9Accuracy (micro-average)
7
May 13, 2026
76.3Accuracy (Exact Match)
7
May 13, 2026
85.1Accuracy
7
May 12, 2026
90.4Accuracy
7
May 12, 2026
74.23Pass@4
7
May 12, 2026
79.3Score
7
May 8, 2026
90.2Score
7
May 8, 2026
0.196Equation-level Diversity
7
May 4, 2026
80Pass@1 Accuracy
7
Apr 28, 2026
93.2Accuracy (SingleEq)
7
Apr 28, 2026
25.6Accuracy
7
Apr 27, 2026
18.9Accuracy (ar)
7
Apr 23, 2026
72.2Accuracy
7
Apr 21, 2026
95.6Accuracy
7
Apr 21, 2026
76.1Accuracy
7
Apr 21, 2026
96.2Accuracy
7
Apr 21, 2026
40Pass@1 (Greedy)
7
Apr 16, 2026
85.6Accuracy
7
Apr 28, 2026
23.3Accuracy
7
Apr 14, 2026
20Accuracy
7
Apr 14, 2026
45.5Accuracy
7
Apr 14, 2026
0.892ASR (Rate)
7
Apr 14, 2026