Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
84.35Avg@8 Accuracy
12
Jun 23, 2026
13,040Reasoning Steps (Tokens)
12
Jun 17, 2026
18,923Thinking Tokens
12
Jun 17, 2026
65.4Avg@32
12
Jun 17, 2026
51.3Avg@32
12
Jun 17, 2026
6,768Response Length (Tokens)
12
Jun 17, 2026
501Response Length (Tokens)
12
Jun 17, 2026
5.83Avg@8
12
Jun 17, 2026
9.17Avg@8
12
Jun 17, 2026
42.19Avg@8
12
Jun 17, 2026
83.85Avg@8
12
Jun 17, 2026
5.83Avg@8
12
Jun 17, 2026
9.17Avg@8
12
Jun 17, 2026
42.19Avg@8
12
Jun 17, 2026
63.25Avg@8 Score
12
Jun 17, 2026
83.85Average Score @8
12
Jun 17, 2026
27.76Avg@8 Accuracy
12
Jun 17, 2026
7.08Avg@8
12
Jun 17, 2026
10.83Avg@8
12
Jun 17, 2026
64.53Avg@8
12
Jun 17, 2026
83.5Avg@8
12
Jun 17, 2026
40.2Accuracy
12
Jun 16, 2026
83.9Accuracy
12
Jun 16, 2026
45.1Average Score @32
12
Jun 16, 2026
65.17Average@32
12
Jun 16, 2026