Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
87.3MATH1 Score
21
Mar 11, 2026
99.49Accuracy
21
May 12, 2026
93.7Accuracy
21
Mar 4, 2026
72.7Average Score
21
Apr 20, 2026
73.95Mean@8 Accuracy
21
Mar 31, 2026
61.2Accuracy (%)
21
Feb 26, 2026
38.6Pass@1 Rate
21
Mar 4, 2026
58.1Accuracy
21
Feb 26, 2026
41.93Performance
21
Feb 26, 2026
73.45Avg@32
21
Feb 26, 2026
54.8Avg@16
21
Jun 9, 2026
60.9Avg@8
21
Mar 10, 2026
87Accuracy
21
Apr 7, 2026
94.92Accuracy
21
Feb 26, 2026
82.63Accuracy
21
Feb 26, 2026
50.84Avg@10 Score
21
Feb 26, 2026
13.67Avg@10
21
Feb 26, 2026
15.33Avg@10 Recall
21
Feb 26, 2026
94.31Accuracy
21
Feb 26, 2026
45.2Accuracy
21
Feb 26, 2026
22Pass@1
21
Feb 26, 2026
62.4Accuracy @ 128
21
Feb 26, 2026
93.1Accuracy @ 128 Tokens
21
Feb 26, 2026
56.79GSM8K Accuracy
21
Feb 26, 2026
71.1Exact Match Accuracy
20
Jun 29, 2026