Long-context performance
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
16GovReport Score
13
Mar 17, 2026
25.2Score
9
Jul 2, 2026
98Score
9
Jul 2, 2026
67.4Score
8
Jul 7, 2026
69Score
5
Jul 2, 2026
99.7Score
5
Jul 2, 2026
89.4Score
5
Jul 2, 2026
64Overall Score
4
Jul 2, 2026
100Score
4
Jul 2, 2026
79Score
4
Jul 2, 2026