Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
1,206# Traces
20
Feb 26, 2026
911Trace Count
20
Feb 26, 2026
28.3Pass@8 Rate
20
Feb 26, 2026
31.3Pass@8
20
Feb 26, 2026
52.5Pass@8
20
Feb 26, 2026
46.67Accuracy (pass@1)
20
Apr 21, 2026
36.72Pass@8
20
Apr 15, 2026
55.1Accuracy
20
May 12, 2026
58.12Overall Score
20
Feb 26, 2026
53.33Avg@8 Score
20
Feb 26, 2026
0.863Composite Score
20
Feb 26, 2026
88.16Composite Score
20
Feb 26, 2026
84.27MATH Accuracy
20
Feb 26, 2026
38.79avg@16 Accuracy
20
Mar 25, 2026
70.97Accuracy (%)
20
Feb 26, 2026
0.986Accuracy
20
Feb 26, 2026
64Accuracy
20
Feb 26, 2026
84Success Rate
20
May 26, 2026
20,417Throughput (tok/s)
20
Feb 26, 2026
38.24Accuracy
20
Feb 26, 2026
73Pass Rate
20
Feb 26, 2026
96Pass Rate
20
Feb 26, 2026
67.8Pass Rate (Avg@8)
20
Feb 26, 2026
20Pass Rate (Avg@8)
20
Feb 26, 2026
76.4Accuracy
20
Feb 26, 2026