Mathematical Problem Solving
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
84.4Accuracy
4
Feb 26, 2026
96Accuracy
3
Jun 29, 2026
31.66Accuracy (Minerva)
3
Jun 23, 2026
12.9Pass@1
3
Jun 17, 2026
30.3Pass@1
3
Jun 17, 2026
58.5AIME Accuracy
3
May 1, 2026
24.2AIME Success Rate
3
Apr 21, 2026
13ECE
3
Mar 5, 2026
0.075ECE
3
Mar 5, 2026
13.2ECE
3
Mar 5, 2026
73.89Accuracy
3
Feb 26, 2026
49.4Average Accuracy
3
Feb 26, 2026
14Baseline Score
2
Jun 2, 2026
45Avg@32
2
Feb 26, 2026
24.6Average Score (@32)
2
Feb 26, 2026
0.668Avg@32
2
Feb 26, 2026
93.4Avg@32
2
Feb 26, 2026
45.3Avg@32
2
Feb 26, 2026
63.5Avg@32
2
Feb 26, 2026
81.3maj@k
2
Feb 26, 2026
66Winrate
1
Jun 23, 2026
89Winrate
1
Jun 23, 2026
45Winrate
1
Jun 23, 2026
2,747N
1
Jun 2, 2026
546N
1
Jun 2, 2026