Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
21.83Accuracy
8
Jul 1, 2026
37.93Accuracy
8
Jul 1, 2026
94.8Accuracy
8
Jun 30, 2026
66.71Average Score
8
Jun 30, 2026
29.5Mean@30
8
Jun 29, 2026
27.6Mean Score @30
8
Jun 29, 2026
26.8Mean Score @30
8
Jun 29, 2026
35.1Mean @30
8
Jun 29, 2026
2.78Speedup
8
Jun 26, 2026
60.1MathVista Score
8
Jun 26, 2026
4.9Mean Accuracy
8
Jun 25, 2026
56Mean Accuracy (MATH500)
8
Jun 25, 2026
61.5Average Accuracy (Pass@1)
8
Jun 23, 2026
90.9Pass@1
8
Jun 23, 2026
41.3Best-of-K Accuracy
8
Jun 23, 2026
92.4Accuracy
8
Jun 15, 2026
67.5Accuracy
8
Jun 12, 2026
51.6GSM8K Score
8
Jun 11, 2026
50.8Exact Match Accuracy
8
Jun 9, 2026
43.3Exact Match Accuracy
8
Jun 9, 2026
56.7Exact Match Accuracy
8
Jun 9, 2026
67.33Pass@1 Accuracy
8
Jun 9, 2026
75.2Majority Score (Avg)
8
Jun 9, 2026
95Maj@16
8
Jun 9, 2026
90Maj@8
8
Jun 9, 2026