List your company

AgentDojo, AgentDyn, SafeToolBench, VPIBench, VisualWebArena, DocVQA

Benchmarks

TaskDatasetSOTA resultTrendResults
AgentDojo, AgentDyn, SafeToolBench, VPIBench, VisualWebArena, DocVQA Stratified 500-task
1,103Unsup. Critical Failures Count
3