Web agent task completion
Benchmarks
DatasetSOTA methodMetricTrendResultsLast updated
42.9Shopping Success Rate
18
May 21, 2026
37.6Accuracy
15
Jun 8, 2026
51.5Accuracy
15
Jun 8, 2026
70.9Accuracy
14
Jun 8, 2026
79Accuracy
14
Jun 8, 2026
90.4Success Rate
4
Feb 26, 2026
0.479Success Rate
4
Feb 26, 2026