Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
6.6Accuracy@16
10
Jun 9, 2026
15.4AIME avg@16 Accuracy
10
Jun 9, 2026
14.8Accuracy@16
10
Jun 9, 2026
31Avg@16 Accuracy
10
Jun 9, 2026
4.28ECE
10
Jun 5, 2026
62.04Average Accuracy
10
Jun 4, 2026
44.73Accuracy (Avg)
10
Jun 4, 2026
49.17Average Accuracy
10
Jun 4, 2026
47.78Average Accuracy
10
Jun 4, 2026
82.5Average Accuracy
10
Jun 4, 2026
92.07Average Accuracy (MATH500)
10
Jun 4, 2026
65.6Accuracy
10
Jun 4, 2026
46.1Average@16
10
Jun 4, 2026
93Accuracy
10
Jun 2, 2026
96.13Accuracy
10
Jun 2, 2026
75.52Pass@1
10
Jun 2, 2026
36.6Avg@8
10
Jun 2, 2026
11.93Average Score @8
10
Jun 2, 2026
89.38Average Score (k=8)
10
Jun 2, 2026
60Accuracy
10
Jun 2, 2026
85.7Accuracy
10
Jun 2, 2026
0.491Loss
10
Jun 1, 2026
79Throughput (TPS)
10
Jun 1, 2026
45.3Average@8 Score
10
May 29, 2026
49.8Average@8 Score
10
May 29, 2026