Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
0Cost (cents)
11
Mar 24, 2026
0Cost (cents)
11
Mar 24, 2026
73.9Accuracy
11
Mar 4, 2026
73.3Accuracy
11
Mar 4, 2026
76.52QA Accuracy
11
Mar 4, 2026
1.89ECE
11
Mar 4, 2026
18.9ROUGE-1
11
Feb 27, 2026
41.8F1 (Berant)
11
Feb 27, 2026
80.42WikiMultiHopQA Pass@1
11
Feb 26, 2026
28.92Accuracy (Normalized)
11
Feb 26, 2026
79ARC Accuracy
11
Feb 26, 2026
0.92Accuracy
11
Feb 26, 2026
63Accuracy
11
Feb 26, 2026
77.31Accuracy (Baseline)
11
Feb 26, 2026
77.3TriviaQA Score
11
Feb 26, 2026
31.31LasJ
11
Feb 26, 2026
33.47LasJ
11
Feb 26, 2026
56.32LasJ
11
Feb 26, 2026
61.59LASJ
11
Feb 26, 2026
36.5LasJ
11
Apr 7, 2026
61.3EM
11
Feb 26, 2026
60.13Accuracy
11
Feb 27, 2026
38.41Total F1
11
Feb 26, 2026
76.56Normalized sub-EM Accuracy
11
Feb 26, 2026
75Normalized Sub-EM Accuracy
11
Feb 26, 2026