Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
13.33Accuracy (AIME 2023)
5
Jul 1, 2026
18.27Accuracy
5
Jul 1, 2026
52.1Pass@1
5
Jun 25, 2026
76.4Pass@1
5
Jun 25, 2026
93.21Accuracy (Source Lib.)
5
Jun 19, 2026
39.6Accuracy
5
Jun 17, 2026
79.38Accuracy (GSM8K)
5
Jun 17, 2026
0.985Sel-AUC
5
Jun 17, 2026
585Score
5
Jun 16, 2026
87.6MTP Acceptance Rate
5
Jun 11, 2026
82.5Pass@1 Rate
5
Jun 11, 2026
1,163Latency (NoKV) (ms)
5
Jun 11, 2026
71.9T Score
5
Jun 11, 2026
93.7Pass@8 Accuracy
5
Jun 9, 2026
99Threshold Accuracy
5
Jun 9, 2026
117.3Threshold Score
5
Jun 9, 2026
86.2Reward
5
Jun 8, 2026
88.9Reward
5
Jun 8, 2026
48.92Accuracy
5
Jun 8, 2026
42.3Pass@1
5
Jun 5, 2026
79.5Avg@16 Accuracy
5
Jun 5, 2026
61.6Accuracy
5
Jun 4, 2026
53.55TPS
5
Jun 4, 2026
79.2Pass@1
5
Jun 2, 2026
62.7Pass@1
5
Jun 2, 2026