Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
93.2ACC (maj@8)
18
Apr 16, 2026
33.33ACC/maj@8
18
Apr 16, 2026
43.3ACC/maj@8
18
Apr 16, 2026
66.78ACC
18
Apr 16, 2026
57.73Accuracy
18
Apr 16, 2026
71.23Accuracy (ACC)
18
Apr 16, 2026
89.58Accuracy (MATH500)
18
Apr 16, 2026
35Accuracy
18
Apr 16, 2026
84.8GSM Accuracy
18
Apr 15, 2026
72.1Accuracy
18
Apr 15, 2026
39.37Accuracy (%)
18
Apr 13, 2026
95Accuracy
18
Apr 8, 2026
95.1Accuracy
18
Apr 8, 2026
93.1Accuracy (Strict)
18
Apr 8, 2026
62.45Math Accuracy
18
Mar 24, 2026
92.27Pass@1
18
Mar 24, 2026
3,636.1Throughput
18
Mar 20, 2026
1,499.6Throughput
18
Mar 20, 2026
30.3Pass@1
18
Mar 18, 2026
54.4Pass@1
18
Mar 18, 2026
58.9Pass@1 Rate
18
Mar 18, 2026
51.5Accuracy
18
Mar 9, 2026
24.7Pass@1
18
Mar 4, 2026
13.3Pass@1
18
Mar 4, 2026
52.3Pass@1
18
Mar 4, 2026