LongMemEval
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
LongMemEval (session-level)
80Ra@5
8
LongMemEval_M session-level granularity binary all-or-nothing recall
80Recall@5
8
LongMemEval-s
60.24o-J Score
8
LongMemEval 1.0 (test)
92.6Overall Accuracy
8
LongMemEval session-level (test)
0.649F1 Score
8
LongMemEval (test)
91.43SS-User Accuracy
7
LongMemEval S (full)
0.462Judgment Score (Jdg)
7
LongMemEval-S session granularity v1
98.7R@5
7
LongMemEval 1.0 (test)
49CSM
7
LongMemEvalS
78.7SSU F1
7
LongMemEval v0.6.0 (strict)
92Accuracy
7
LongMemEval single run
46.7F1 Score
7
LongMemEval S (test)
0.195F1 Score
7
LongMemEval S (test)
6.61F1 Score
7
LongMemEval S (test)
15.03F1 Score
7
LongMemEval S (test)
14.14F1 Score
7
LongMemEval-M
8.67F1 Score
7
LongMemEval-M
6.23F1 Score
7
LongMemEval-M
4.81F1 Score
7
LongMemEval-M
12.69F1 Score
7
LongMemEval-M
10.88F1
7
LongMemEval-M
13.79F1 Score
7
LongMemEval S
11.83F1 Score
7
LongMemEval-S
17.92F1 Score
7
LongMemEval
2.56Storage Size (GiB)
6