Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
128SFT Data Used
9
May 15, 2026
70Accuracy
9
May 15, 2026
73.8BCA
9
May 13, 2026
5.63SpeedUp
9
May 12, 2026
5.28SpeedUp
9
May 12, 2026
9.78Pass@1
9
May 12, 2026
1.69Pass@1
9
May 12, 2026
20.5Pass@1
9
May 12, 2026
44Pass@1
9
May 12, 2026
4.39Pass@1
9
May 12, 2026
41.74Pass@1
9
May 12, 2026
12.31Exact Match Accuracy
9
May 11, 2026
9.2Pass@1
9
May 11, 2026
12Pass@1
9
May 11, 2026
53.9Pass@1
9
May 11, 2026
97Accuracy
9
May 7, 2026
69.4Accuracy
9
May 6, 2026
90.33Accuracy (SVAMP)
9
May 6, 2026
52.69Accuracy
9
May 6, 2026
89.31Accuracy (GSM8K)
9
May 6, 2026
61.67Accuracy
9
May 1, 2026
94.8Accuracy
9
May 1, 2026
48.3Accuracy
9
Apr 28, 2026
0.825Pass@1
9
Apr 28, 2026
0.5Pass@1 Accuracy
9
Apr 28, 2026