Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
88.68pass@1
30
Feb 26, 2026
40.66EM
30
Apr 28, 2026
62.03TruthfulQA Score
30
Mar 4, 2026
83.9Accuracy
30
Mar 13, 2026
94.1F1 (de)
30
Apr 21, 2026
60RACC
29
May 25, 2026
15.92F1 Score
29
Jun 26, 2026
41.8Accuracy (Normalized)
29
May 28, 2026
59Accuracy
29
Feb 26, 2026
48.8Exact Match
29
Jun 4, 2026
79.49ECQA Accuracy
29
Feb 26, 2026
28.9Unreliability Rate
28
Jun 30, 2026
59.1Unreliability Rate
28
Jun 30, 2026
41.7Unreliability Rate
28
Jun 30, 2026
41.1Unreliability Rate
28
Jun 30, 2026
38.8Unreliability Rate
28
Jun 30, 2026
53.52EM
28
May 12, 2026
76F1 Score
28
May 8, 2026
62.1Utility Score
28
Apr 24, 2026
70.1Utility
28
Apr 24, 2026
54.1Utility
28
Apr 24, 2026
98.3StrQA Accuracy
28
Apr 23, 2026
79.2HellaSwag Accuracy
28
Apr 21, 2026
50.63Average Accuracy
28
Apr 8, 2026
0.633Overall Accuracy
28
Apr 8, 2026