Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
74.8Accuracy
45
Apr 28, 2026
94.3Accuracy
44
Jul 9, 2026
83.2Accuracy
44
Feb 26, 2026
95.5P@1
43
Jun 9, 2026
89.6EM Accuracy
43
Jun 24, 2026
67.56Accuracy
42
Jul 3, 2026
53.96Accuracy
42
Apr 21, 2026
89.27Score
42
Jun 19, 2026
31.1Accuracy
40
Jun 9, 2026
70Accuracy
40
Feb 26, 2026
76.9Accuracy
40
Feb 26, 2026
54.22Accuracy
39
Mar 10, 2026
72.6ARC-e Accuracy
38
May 26, 2026
82.8Avg@32
38
May 11, 2026
63.6Accuracy
37
May 20, 2026
84OCR Score
37
May 5, 2026
100Accuracy (BigBenchHard)
37
Jul 7, 2026
77.5Accuracy
36
Jun 12, 2026
81.35Accuracy
36
May 25, 2026
58.08Accuracy
36
Jun 2, 2026
78MMLU-Pro Reasoning Score
36
Jun 2, 2026
85.5GPQA Diamond Reasoning Score
36
May 19, 2026
92.87Accuracy (Natural Sciences)
36
May 19, 2026
59.44Average Score
36
Mar 17, 2026
82Accuracy
35
Mar 4, 2026