Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
54.4Accuracy
14
Jul 7, 2026
88.7Accuracy
14
Jul 7, 2026
83.1Accuracy
14
Jul 2, 2026
64.1Overall Score (ALL)
14
Jun 25, 2026
8.78Speedup
14
Jun 18, 2026
9.64Speedup
14
Jun 18, 2026
7.82Speedup
14
Jun 18, 2026
77.6Accuracy (greedy pass@1)
14
Jun 17, 2026
13Normalized Spearman Distance
14
Jun 17, 2026
26.76Mean Accuracy
14
Jun 9, 2026
66.72Mean Accuracy (MATH 500)
14
Jun 9, 2026
66.69Accuracy (Mean)
14
Jun 9, 2026
47Accuracy (Olympiad)
14
Jun 9, 2026
79.7Accuracy
14
Jun 9, 2026
65.4Accuracy
14
Jun 9, 2026
12.3Accuracy
14
Jun 9, 2026
96.51Accuracy
14
Jun 5, 2026
73.33Accuracy
14
Jun 5, 2026
75Accuracy
14
Jun 5, 2026
96.81Accuracy
14
Jun 5, 2026
70Accuracy
14
Jun 5, 2026
73.33Accuracy
14
Jun 5, 2026
80AIME25 Accuracy
14
Jun 5, 2026
24.5AIME 2024 (avg@32)
14
Jun 4, 2026
96.58Accuracy
14
Jun 4, 2026