Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
94.47Pass@4 Score
27
May 12, 2026
94.2Pass@1 Accuracy
27
Jun 11, 2026
71.88Accuracy
27
Apr 14, 2026
94.77Accuracy
27
Apr 14, 2026
96.7Accuracy
27
Apr 7, 2026
72Top-1 Accuracy
27
Jun 9, 2026
73.8Accuracy
27
Jun 9, 2026
94.6Pass@1
27
May 11, 2026
47.79Pass@1
27
Mar 17, 2026
50.07Pass@1
27
Mar 17, 2026
83.8Pass@1
27
Mar 17, 2026
79.52Acc@8
27
Apr 16, 2026
69.32Weighted Majority Voting Accuracy
27
Mar 4, 2026
83.33Weighted Accuracy
27
Mar 4, 2026
46.88Accuracy Pass@1
27
May 11, 2026
91.3Flex Match
27
Feb 27, 2026
92.16GSM Accuracy
27
Jun 1, 2026
80.74Accuracy (@avg1)
27
Mar 6, 2026
61Accuracy
27
Feb 26, 2026
68.44Avg@32
27
Feb 26, 2026
86.75Avg@32
27
Feb 26, 2026
11.8Mean@64 Accuracy
27
Feb 26, 2026
64.1Accuracy
27
Feb 26, 2026
58.15GSM8k Accuracy
27
Feb 26, 2026
89.7Accuracy
27
Apr 23, 2026