Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
45.2Score
4
May 18, 2026
33.8Accuracy (128 tokens)
4
May 15, 2026
74.76Accuracy (128 tokens)
4
May 15, 2026
88.9Accuracy (Hard Math Stream)
4
May 13, 2026
63.3REST Score
4
May 12, 2026
73.3REST Score
4
May 12, 2026
70REST Score
4
May 12, 2026
76.7REST Score
4
May 12, 2026
80REST
4
May 12, 2026
41.2REST Score
4
May 12, 2026
5.12SGLang Acceptance Length
4
May 12, 2026
4.42Acceptance Length
4
May 12, 2026
5.89Acceptance Length
4
May 12, 2026
5.73Acceptance Length
4
May 12, 2026
6.8Avg@4
4
May 12, 2026
14.4Avg@16 Score
4
May 12, 2026
91.43Accuracy
4
May 11, 2026
69.5Accuracy (%)
4
May 11, 2026
94Accuracy
4
May 11, 2026
97.2Accuracy
4
May 11, 2026
94.8Accuracy
4
May 11, 2026
50.29Pass@8
4
May 8, 2026
23.33AIME 24 Pass@8 Score
4
May 8, 2026
74.53Pass@8
4
May 8, 2026
82.6Pass@8
4
May 8, 2026