Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
40.8Accuracy (HLE Reasoning)
63
Apr 23, 2026
87.3Average Accuracy
62
May 19, 2026
94.21Average Score
60
Jul 3, 2026
82.3Accuracy
60
Jun 29, 2026
96Pass@1
60
Jun 4, 2026
84.61Accuracy
60
May 18, 2026
83.5Accuracy
58
Jun 30, 2026
70Accuracy
58
Mar 10, 2026
91.35Accuracy
57
Jun 11, 2026
92Accuracy
57
Mar 4, 2026
0.946Accuracy (ARC)
56
Jun 23, 2026
98.8Accuracy (GSM8K)
55
May 27, 2026
84.5BBH Score
53
Jun 4, 2026
59.47Accuracy
53
May 12, 2026
74.99BBH Accuracy
51
Jul 7, 2026
81.1Score
51
Jul 3, 2026
65.69BBH 3-shot Score
49
Apr 10, 2026
83.2Accuracy
49
Jun 2, 2026
83.69BBH Pass@1
49
May 27, 2026
85.57Pass@1
48
May 12, 2026
73Accuracy
46
Mar 24, 2026
48.82Reasoning Score (0-shot)
46
Mar 4, 2026
0.85AUROC
46
Apr 23, 2026
97.67Accuracy
46
May 28, 2026
85.8Accuracy
45
Apr 8, 2026