Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
85Accuracy
29
Feb 26, 2026
45.6Average Score
29
Apr 2, 2026
97.3During-task Accuracy (MATH 500)
29
Feb 26, 2026
93.4Pass@1
29
Mar 24, 2026
74.04Pass@1
29
May 21, 2026
40Pass@8
29
Feb 26, 2026
77.6Avg@16
29
Feb 26, 2026
93.9Accuracy
29
Feb 26, 2026
53.6Accuracy
29
Feb 26, 2026
76.25AIME24 Score
28
Jun 23, 2026
29.76Accuracy
28
Jun 18, 2026
32.8Accuracy
28
Jun 4, 2026
16.6Accuracy
28
Jun 4, 2026
49.8Accuracy
28
Jun 4, 2026
26Accuracy
28
Jun 4, 2026
6.76Speedup
28
Jun 30, 2026
81.6Easy Accuracy
28
Jun 2, 2026
70Mean@8
28
Jun 2, 2026
85.3GSM8K Accuracy
28
May 29, 2026
57Exact Match
28
May 19, 2026
70Exact Match
28
May 19, 2026
42.53Accuracy (Minerva 272)
28
Apr 30, 2026
56.85Accuracy
28
Apr 30, 2026
81.7Accuracy
28
Apr 28, 2026
80.9Math500 Score
28
May 29, 2026