Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
78.3Accuracy
11
Apr 21, 2026
8.67ECE
11
Mar 5, 2026
7.26ECE
11
Mar 5, 2026
0.061ECE
11
Mar 5, 2026
6.66ECE
11
Mar 5, 2026
32.17Accuracy
11
Mar 4, 2026
69.48PASS@1
11
Apr 1, 2026
55Accuracy (avg@1)
11
Mar 4, 2026
93.8Accuracy (avg@32)
11
Jun 4, 2026
89.2Accuracy (avg@1)
11
Mar 4, 2026
64.39Average Accuracy
11
Feb 26, 2026
62.7Accuracy
11
Jun 9, 2026
81.67Accuracy
11
Apr 21, 2026
55.5Best-of-8 Accuracy
11
May 12, 2026
94.49Accuracy
11
Feb 26, 2026
91MATH500
11
Feb 26, 2026
93.6Average Score (avg@1)
11
Feb 26, 2026
63.3Avg@8
11
Feb 26, 2026
53.8Avg@8 Score
11
Feb 26, 2026
80AUCOAA
11
Feb 26, 2026
81.8AUCOAA
11
Feb 26, 2026
91.9AUCOAA
11
Feb 26, 2026
25.6Accuracy
11
Feb 26, 2026
29.2Accuracy
11
Feb 26, 2026
80.6Accuracy
11
Feb 26, 2026