List your company

Agent Task Completion and Safety on AgentDojo Slack stock (test)

68.6UtilityCaMeL
38.85646.57854.362.022May 27, 2026

Evaluation results

MethodLinks
CaMeLModel=o4-mini-high2026.0568.60
LACUNAModel=o4-mini-high2026.0549.521
CaMeLModel=gemini-2.5-pro2026.0548.60
TACITModel=o4-mini-high2026.0547.620
LACUNAModel=gemini-2.5-pro2026.0542.862
TACITModel=gemini-2.5-pro2026.05400