Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
68.8Accuracy @73.7%
9
Mar 20, 2026
90Accuracy
9
Jun 1, 2026
43.8Acc @ First
9
Mar 19, 2026
16.3Accuracy @ First Answer
9
Mar 19, 2026
47.6Accuracy @ First Step
9
Mar 19, 2026
76.8Accuracy @ First Step
9
Mar 19, 2026
41.41Accuracy
9
Mar 17, 2026
78.47Math Score
9
Mar 12, 2026
74.2Pass@1
9
Mar 11, 2026
99.2Final Accuracy
9
Mar 10, 2026
66.6Standard Correctness (SC)
9
Mar 4, 2026
47.1Pass@1 Accuracy
9
Mar 4, 2026
80Pass@1 Accuracy
9
Mar 4, 2026
70Pass@1 Accuracy
9
Mar 4, 2026
20Pass@1 Accuracy
9
Mar 4, 2026
43.3Pass@1 Accuracy
9
Mar 4, 2026
67.5Pass@1 Accuracy
9
Feb 27, 2026
85.2Pass@1 Accuracy
9
Feb 27, 2026
92Accuracy
9
Feb 27, 2026
81.67Mean@8 Accuracy
9
Feb 26, 2026
47.23EM Accuracy
9
Feb 26, 2026
86.02Avg Accuracy
9
Feb 26, 2026
52.98Avg Acc (A_bar)
9
Feb 26, 2026
30.63Accuracy
9
Feb 26, 2026
47.06Mean @16
9
Mar 4, 2026