Agentic Task Performance
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
96.4CAP
48
May 13, 2026
91.4Completion Accuracy (CAP)
48
May 13, 2026
56.7Average Score
16
Jul 8, 2026
52Pass@1 Success Rate
16
Apr 29, 2026
45Pass^1 Rate
16
Apr 29, 2026
90.4Success Rate
15
Mar 12, 2026
65.8Easy Accuracy
6
Feb 26, 2026
92.3Information Lookup Success Rate
4
May 1, 2026