Long-context retrieval and tracking
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
88.94S-N Score
15
May 1, 2026
89.4S-N Score
15
May 1, 2026
99.54S-N Score
15
May 1, 2026
85.45S-N Score
10
May 1, 2026
79.21S-N
10
May 1, 2026
89.09S-N Score
10
May 1, 2026
83.01FullKV
2
Jun 16, 2026