Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
256Mean Tokens Used
8
Mar 20, 2026
512Token Count
8
Mar 20, 2026
512Token Count
8
Mar 20, 2026
72Accuracy
8
Mar 18, 2026
94Accuracy
8
Mar 18, 2026
86.7Accuracy
8
Mar 18, 2026
66.9Accuracy
8
Mar 17, 2026
53.2Accuracy
8
Mar 17, 2026
0.839Accuracy
8
Mar 17, 2026
61Accuracy
8
Mar 17, 2026
43.8Accuracy
8
Mar 17, 2026
80.2Accuracy
8
Mar 17, 2026
49.37F1 Score
8
Mar 12, 2026
28.64F1 Score
8
Mar 11, 2026
80.22F1 Score (SQuAD v1.1)
8
Mar 10, 2026
4.6Relevance
8
Mar 5, 2026
53.37BLEU Score
8
Mar 5, 2026
85.2Best Accuracy
8
Mar 4, 2026
23.63F1 Score
8
Feb 27, 2026
29.71F1 Score
8
Feb 27, 2026
56.08F1
8
Feb 27, 2026
40.17F1 Score
8
Feb 27, 2026
53TruthfulQA Accuracy
8
Feb 26, 2026
6.2Performance Score
8
Feb 26, 2026
49.2Performance Score
8
Feb 26, 2026