Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
89.7EM
4
Feb 26, 2026
28.57SACC
4
Feb 26, 2026
61.18SACC
4
Feb 26, 2026
52SACC
4
Feb 26, 2026
93.8F1 Score
4
Feb 26, 2026
36Aggregate Precision
4
Feb 26, 2026
46.41EM
4
Feb 26, 2026
45.12Exact Match (EM)
4
Feb 26, 2026
45.9Hits@1
4
Feb 26, 2026
35Accuracy
4
Feb 26, 2026
39.4Accuracy
4
Feb 26, 2026
38.5Accuracy
4
Feb 26, 2026
60.31Exact Match (EM)
4
Feb 26, 2026
46.68EM
4
Feb 26, 2026
50.51Exact Match (EM)
4
Feb 26, 2026
40.9F1 Score
4
Feb 26, 2026
31.3Top-1 Accuracy
4
Feb 26, 2026
42.3Average Accuracy
3
Jul 7, 2026
64.3Accuracy
3
Jul 3, 2026
78.6Accuracy
3
Jul 3, 2026
63.54QA Accuracy (8K Context)
3
Jul 3, 2026
63.2F1 Score
3
Jul 3, 2026
68.5Always-RAG Accuracy
3
Jun 30, 2026
78.5Always-RAG Accuracy
3
Jun 30, 2026
75.4Accuracy
3
Jun 19, 2026