Zero-shot Task Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
68.58Accuracy
83
Jun 4, 2026
69.48Accuracy
28
May 12, 2026
68.660-shot Average
26
Mar 18, 2026
54.7Average Accuracy
10
Feb 26, 2026
35OpenQ (Zero-shot)
4
May 28, 2026