Multimodal Agent Task
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
23.85Average@4
40
May 12, 2026
46.5Average@4
40
May 12, 2026
43.72Average@4
24
Mar 13, 2026
47.75Average@4
24
Mar 13, 2026
47.17Success Rate (Level 1)
10
Jun 12, 2026