Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
86.8Exact Match Accuracy
7
Feb 26, 2026
76.7ASR
7
Feb 26, 2026
95.4Pass@1
7
Feb 26, 2026
97.1Average@32
7
Feb 26, 2026
25.5Average@4
7
Feb 26, 2026
66.3Average@4
7
Feb 26, 2026
7.5Average@32
7
Feb 26, 2026
41.72Accuracy
7
Feb 26, 2026
64.4Avg@6
7
Feb 26, 2026
70.6Avg@6
7
Feb 26, 2026
83.3Avg@6
7
Feb 26, 2026
46.7Accuracy (%)
7
Feb 26, 2026
21.56Mean@4
7
Feb 26, 2026
92Score
7
Feb 26, 2026
90.9Accuracy
7
Feb 26, 2026
80.9Accuracy
7
Feb 26, 2026
96.2Accuracy (Baseline)
7
Feb 26, 2026
96.2Mean@16
6
Jul 8, 2026
43.35CMath Score
6
Jul 7, 2026
19.8Accuracy
6
Jul 7, 2026
66.5Accuracy
6
Jul 7, 2026
43.45Accuracy
6
Jul 2, 2026
80Pass@1 Accuracy
6
Jul 1, 2026
95.6Pass@1 Accuracy
6
Jul 1, 2026
80Pass@1 Accuracy
6
Jul 1, 2026