Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
0.955p0 Score
3
Apr 21, 2026
44.5Detection Accuracy
3
Apr 21, 2026
50.81TPS
3
Apr 21, 2026
69.95TPS
3
Apr 21, 2026
43Score
3
Apr 20, 2026
62.9Average Score
3
Apr 20, 2026
48.6Overall Accuracy
3
Apr 20, 2026
62.55GSM8k Zero-shot Accuracy
3
Apr 20, 2026
96.8Accuracy (en)
3
Apr 14, 2026
40Accuracy
3
Apr 13, 2026
93.33Exact Match (EM)
3
Apr 10, 2026
95.3Strict Accuracy
3
Apr 8, 2026
13.6Peak Accuracy
3
Apr 7, 2026
82.6Peak Accuracy
3
Apr 7, 2026
15.6AIME24 Score
3
Apr 7, 2026
43.7Recovery Rate
3
Apr 6, 2026
72.2Recovery Rate
3
Apr 6, 2026
39.6Recovery Rate
3
Apr 6, 2026
37.3Recovery Rate
3
Apr 6, 2026
40Recovery Rate
3
Apr 6, 2026
31Recovery Rate
3
Apr 6, 2026
29.7Recovery Rate
3
Apr 6, 2026
27.6Recovery Rate (%)
3
Apr 6, 2026
78.09Accuracy
3
Mar 18, 2026
47.5Accuracy
3
Mar 18, 2026