Long-context Memory Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
95.6Average Score
103
Jun 2, 2026
81F1 Score
35
Apr 2, 2026
61.56Single-Hop F1
21
Jul 2, 2026
75Overall Score
12
Feb 26, 2026
87.5Recall
10
May 15, 2026
89.6Accuracy
9
Apr 27, 2026
92Accuracy
7
May 7, 2026
63.07Accuracy
6
Apr 14, 2026
39.39SingleHop F1
3
Jul 2, 2026
0.01Delta
1
Jun 11, 2026
0.1Delta
1
Jun 11, 2026
0.05Delta Score
1
Jun 11, 2026
0.06Delta
1
Jun 11, 2026
0.08Delta
1
Jun 11, 2026