Knowledge
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
30.06Score
10
Feb 26, 2026
12.92Score
9
Jun 16, 2026
31.5Score
9
Jun 16, 2026
76.8Score
9
Jun 16, 2026
36.36GPQA Score
9
Apr 8, 2026
71.9EM
9
Mar 20, 2026
77.9Knowledge EM
9
Mar 20, 2026
57.5GPQA Score
9
Feb 27, 2026
84.07ARC-C
9
Feb 27, 2026
58.1Accuracy
7
May 27, 2026
26.7Accuracy
7
May 27, 2026
80.9Score
7
Feb 26, 2026
70.9Score
7
Feb 26, 2026
91.7Knowledge Accuracy (MMLU Redux)
6
Jun 1, 2026
0.894Kendall Tau
5
Jun 17, 2026
0.96Kendall's Tau
5
Jun 17, 2026
66.4Score
5
Jun 16, 2026
46.1Accuracy
5
Mar 30, 2026
57.23BLEU-1
4
Mar 11, 2026
65.2Top-1 Score
3
Jun 17, 2026
78.5Top-1 Score
3
Jun 17, 2026
32.92Accuracy (Knowledge, test)
3
May 19, 2026
32.11Knowledge CMMLU Test Accuracy
3
May 19, 2026
30.59Accuracy
3
Feb 27, 2026
75.9Score
3
Feb 26, 2026