Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
69.12S_Math Score
4
Jun 4, 2026
23.6Accuracy (Config 64)
4
Jun 2, 2026
67.6MATH multi@5 Accuracy
4
Jun 1, 2026
66.67Greedy Accuracy
4
May 29, 2026
49.08Pass@32
4
May 29, 2026
95.3SCR+
4
May 29, 2026
39.4NT Accuracy
4
May 28, 2026
69.94Accuracy
4
May 26, 2026
88.1Accuracy (Before GSM8K)
4
May 26, 2026
93.6GSM8K Score
4
May 26, 2026
48.65Pass@4
4
May 26, 2026
78.8Majority@2048
4
May 26, 2026
23.3Majority@2048
4
May 26, 2026
56.67Precision@8
4
May 20, 2026
70P@8
4
May 20, 2026
90.3Accuracy
4
May 20, 2026
55.8Accuracy
4
May 20, 2026
47.9Accuracy
4
May 20, 2026
86.9Accuracy (AMC 23)
4
May 20, 2026
3.04Success Rate (SR)
4
May 20, 2026
3.25Success Rate (SR)
4
May 20, 2026
34.6Accuracy@16
4
May 19, 2026
55.2Acc@16
4
May 19, 2026
27.7Average Score
4
May 18, 2026
42.6Score
4
May 18, 2026