Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
80Accuracy
5
Jun 2, 2026
80.37Pass@1
5
Jun 1, 2026
0.142Loss
5
Jun 1, 2026
46.74Accuracy
5
May 28, 2026
76.96Accuracy
5
May 28, 2026
64.9Pass@1
5
May 26, 2026
13.04Pass@1
5
May 26, 2026
86.7Accuracy
5
May 22, 2026
53Accuracy
5
May 22, 2026
91.3Accuracy
5
May 22, 2026
78.2GSM8K Accuracy
5
May 21, 2026
7.98Average Acceptance Length (τ)
5
May 20, 2026
55.6Score
5
May 18, 2026
86.67AIME24 pass@32
5
May 15, 2026
73.8Accuracy
5
May 15, 2026
91.9Accuracy
5
May 13, 2026
75.1Pass@1 Accuracy
5
May 13, 2026
7.39GSMIR Score
5
May 13, 2026
79.1Math-P Hard Score
5
May 13, 2026
46.38Pass@1
5
May 13, 2026
92.16Pass@1
5
May 13, 2026
50ACC
5
May 13, 2026
89.9Accuracy
5
May 13, 2026
0.06Mean Absolute Difference (|Δs|)
5
May 12, 2026
53.33Pass@8
5
May 12, 2026