Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
95.4Accuracy
770
Jun 29, 2026
94.5Accuracy
269
Jun 19, 2026
92.86Accuracy
264
Jun 11, 2026
96.63Accuracy
242
Jul 9, 2026
91.84HellaSwag Accuracy
209
May 27, 2026
91.9Accuracy
185
May 18, 2026
85.2Accuracy
172
Jun 5, 2026
96.5Accuracy
168
Jun 5, 2026
96.3Accuracy (ARC-c)
113
Jul 2, 2026
1Accuracy
111
Jun 30, 2026
68.77Accuracy
108
Feb 26, 2026
97.2Accuracy
100
May 26, 2026
100Accuracy (%)
94
May 11, 2026
73.9Accuracy
94
May 19, 2026
71.43Accuracy
92
Jun 11, 2026
90.1Pass@1
92
Jun 11, 2026
88.4Accuracy
92
May 20, 2026
96.7Accuracy
81
May 18, 2026
92Accuracy
80
Feb 26, 2026
100Accuracy
72
Apr 8, 2026
55.45pass@1 Accuracy
71
Jun 18, 2026
64.66BBH
66
May 27, 2026
55Average Score
66
Jul 3, 2026
86.3Accuracy on AIME 24
65
May 8, 2026
64.7Score
65
Jul 7, 2026