Mathematics Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
60.54Accuracy
35
May 4, 2026
96.7AIME 2025 Accuracy
18
Jun 16, 2026
91.3Accuracy
18
Jun 8, 2026
80.6Accuracy
15
Jun 16, 2026
97.1Accuracy (AIME 2026)
14
Jun 16, 2026
14.7Accuracy
10
Apr 28, 2026
85Pass@16
10
Feb 26, 2026
78.3Accuracy
6
Apr 3, 2026
89.8Pass@16
6
Feb 26, 2026
98.6Exact Match (EM)
5
Mar 31, 2026
92.8Exact Match (EM)
5
Mar 31, 2026
97.6Accuracy (%)
4
May 20, 2026
96.36Accuracy
4
May 20, 2026