Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
74Accuracy (Clean)
16
Jul 2, 2026
67Clean Accuracy (NQ)
16
Jul 2, 2026
62.9Exact Match
16
Jun 25, 2026
1.57ECE
16
Jun 5, 2026
86Accuracy
16
Jun 15, 2026
86.4Accuracy
16
May 22, 2026
55.5Initial Accuracy
16
May 22, 2026
75.39Throughput
16
May 22, 2026
68.9Correctness (%)
16
Jun 30, 2026
22.6Exact Match
16
May 14, 2026
43.5Exact Match Accuracy
16
May 14, 2026
61Overall Accuracy (MMLU-Pro QA)
16
May 12, 2026
42.4EM
16
May 12, 2026
1.7Average Rank
16
May 11, 2026
37.7Exact Match (EM)
16
May 11, 2026
26.933Accuracy
16
Apr 28, 2026
66.43R@1
16
Apr 27, 2026
27.22Accuracy
16
Apr 23, 2026
56.7Normalized Accuracy
16
Apr 21, 2026
83.5Accuracy
16
Apr 21, 2026
73.86Overall Accuracy
16
Apr 8, 2026
72.34Exact Match Accuracy
16
Apr 8, 2026
96.3C1 Accuracy
16
Apr 8, 2026
63.96Exact Match Accuracy
16
Apr 8, 2026
41.11Exact Match
16
Apr 3, 2026