Open-domain Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
69.6Accuracy
3
May 12, 2026
44.8Accuracy
3
May 12, 2026
35.28BLEU-1
3
Apr 23, 2026
75.23Exact Match (EM)
3
Apr 23, 2026
38.5EM
3
Feb 26, 2026
0.626EM
3
Feb 26, 2026
60.4EM
3
Feb 26, 2026
75.8Accuracy
3
Feb 26, 2026
74.6F1
3
Feb 26, 2026
79.9F1 Score
3
Feb 26, 2026
77LAMA Score
2
Jun 29, 2026
58.47Output Token Throughput (tok/s)
2
Feb 26, 2026
26.6Exact Match
2
Feb 26, 2026
34.2EM
2
Feb 26, 2026
50EM
2
Feb 26, 2026
34.2EM
2
Feb 26, 2026
58.5F1
2
Feb 26, 2026
—
—Primary metric
0
Apr 7, 2026
—
—Accuracy
0
Feb 26, 2026
—
—Accuracy
0
Feb 26, 2026
—
—Acc
0
Feb 26, 2026