Knowledge Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
82.03MMLU Accuracy
64
Jun 2, 2026
83Accuracy
45
Apr 6, 2026
93.02Natural Sciences Score
36
May 19, 2026
71.9Accuracy (IKP)
30
Apr 29, 2026
85.1Accuracy
21
Feb 26, 2026
0.108Brier Score
18
Feb 26, 2026
3.94Accuracy
12
Jun 23, 2026
81.23Accuracy
10
Mar 18, 2026
77.44Accuracy
9
Mar 18, 2026
29.87Accuracy (MMLU_c)
7
Apr 23, 2026
37.8Accuracy
7
Apr 23, 2026
32Accuracy
7
Apr 23, 2026
36.88Accuracy
7
Apr 23, 2026
84.6Accuracy
6
Jul 1, 2026
68.6Accuracy
6
Jul 1, 2026
15.85Accuracy
6
May 12, 2026
11.01Accuracy
6
May 12, 2026
33.49Accuracy
6
May 12, 2026
42.03Accuracy
6
May 12, 2026
32.03Accuracy
6
May 12, 2026
58Accuracy
6
Feb 26, 2026
0.777BPB
6
Feb 26, 2026
11.2Accuracy
6
Feb 26, 2026
49.7Accuracy
5
Mar 17, 2026
40Improvement (%)
5
Feb 26, 2026