Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
66.7Pass@256
10
Apr 14, 2026
63.3Pass@256
10
Apr 14, 2026
98.6Pass@128
10
Apr 14, 2026
91.4Pass@16
10
Apr 14, 2026
43.66Average Accuracy @8
10
Apr 14, 2026
84.68Avg Success Rate (@8)
10
Apr 14, 2026
6.73Latency (s)
10
Apr 13, 2026
67.6Pass@4
10
Apr 3, 2026
93.6pass@4
10
Apr 3, 2026
66.5Exact Match Score
10
Jun 15, 2026
1Total Tokens
10
Apr 2, 2026
96.4Score
10
Apr 1, 2026
96.7Accuracy
10
May 20, 2026
95.3Accuracy (GSM8K)
10
Mar 24, 2026
65.56Accuracy
10
Mar 24, 2026
94.1AUROC
10
Mar 20, 2026
65.83Accuracy
10
Mar 20, 2026
91.56Accuracy
10
Mar 20, 2026
43.33Accuracy
10
Mar 17, 2026
1.2Average Acceptance Length (τ)
10
Mar 11, 2026
4.96Average Acceptance Length
10
Mar 11, 2026
5.41Average Acceptance Length (τ)
10
Mar 11, 2026
79.8Accuracy
10
Mar 10, 2026
88.4Accuracy
10
Mar 10, 2026
96Accuracy
10
Mar 10, 2026