Knowledge
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
85.93Accuracy
171
Jul 2, 2026
80.4Score
98
Jul 7, 2026
90.4Accuracy
51
Mar 31, 2026
81.3Accuracy (GPQA Knowledge)
49
Jul 7, 2026
74.01Knowledge (5-shot) Score
46
Mar 4, 2026
33.03Score
37
Mar 4, 2026
44.65Knowledge Score (5-shot)
37
Mar 4, 2026
69.16Score
35
Mar 4, 2026
69.8Knowledge Accuracy
33
Jun 24, 2026
99.54ARC-E Score
31
Feb 27, 2026
96.43ARC-C Score
31
Feb 27, 2026
36.1MMLU
30
Apr 14, 2026
63.85Accuracy
28
Mar 4, 2026
84.72Knowledge Score
25
May 26, 2026
60.66Score
24
Jun 18, 2026
61.98Accuracy
21
Feb 26, 2026
66.91Score
20
Feb 26, 2026
58.6Accuracy
19
May 27, 2026
92.31Accuracy
19
Jul 3, 2026
0.682C-Eval Knowledge Accuracy
18
May 26, 2026
90Knowledge (MMLU-Redux) Score
17
Jul 7, 2026
88.12Score
17
Apr 14, 2026
60.1MMLU (5-shot)
13
May 27, 2026
77Performance
11
Feb 26, 2026
72.93Score
10
May 27, 2026