Retrieval-Augmented Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
87.17Context Relevance
8
Jul 7, 2026
87.63Context Relevance
8
Jul 7, 2026
86.72Context Relevance
8
Jul 7, 2026
86.1Score
8
Mar 10, 2026
0.902Score
8
Mar 10, 2026
46.24Accuracy
8
Feb 26, 2026
22.25Accuracy
8
Feb 26, 2026
36.2METEOR
8
May 12, 2026
0.464METEOR
8
Feb 26, 2026
77.9Accuracy
8
Feb 26, 2026
0.132Utility Score
7
Jul 9, 2026
54.23HotpotQA Score
7
Apr 24, 2026
63.2Comprehensiveness
6
Jul 8, 2026
4.404CO
6
May 27, 2026
40.3Average F1 Score
6
Mar 25, 2026
83.1Accuracy
6
Feb 26, 2026
3.84#MAT
6
Feb 26, 2026
86Precision@10
5
Jun 11, 2026
66.7Overall Accuracy
5
Apr 23, 2026
95F1 Score
5
Apr 10, 2026
76.3Faithfulness
5
Feb 26, 2026
73.9Faithfulness
5
Feb 26, 2026
82.9Faithfulness
5
Feb 26, 2026
68.8Faithfulness
5
Feb 26, 2026
76Faithfulness
5
Feb 26, 2026