Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
70.81EM
10
Feb 26, 2026
9.54F1 Score
10
Feb 26, 2026
27.61F1
10
Feb 26, 2026
51.64F1
10
Feb 26, 2026
66.86F1 Score
10
Feb 26, 2026
28.05Accuracy
10
Feb 26, 2026
84.1EM
10
Feb 27, 2026
37.05EM
10
Feb 26, 2026
0.612Execution Accuracy
10
Feb 27, 2026
90.9Accuracy
10
Jul 8, 2026
80F1 Score
10
Feb 26, 2026
72BLEU-1
10
Feb 26, 2026
0.891Score
10
Mar 10, 2026
81.4EM
10
Feb 26, 2026
21Mean Error Rate
10
Feb 26, 2026
86.3EM
10
Feb 27, 2026
0.4Task 1 Accuracy
10
Feb 26, 2026
83.3Accuracy
9
Jun 30, 2026
62.93Exact Match (EM)
9
Jun 19, 2026
77.03Exact Match (EM)
9
Jun 19, 2026
66.38ARC-C Utility
9
Jun 16, 2026
64Accuracy (Clean)
9
Jun 16, 2026
54.52Accuracy (Clean)
9
Jun 16, 2026
0.01Latency p50 (s)
9
Jun 11, 2026
8.81Quality Score
9
Jun 4, 2026