List your company

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

74.3UtilityCaMeL
-1.51618.16737.8557.533May 27, 2026

Evaluation results

MethodLinks
CaMeLModel=o4-mini-high2026.0574.30
LACUNAModel=o4-mini-high2026.0571.430
TACITModel=o4-mini-high2026.0568.57100
TACITModel=gemini-2.5-pro2026.0564.290
LACUNAModel=gemini-2.5-pro2026.0563.57100
CaMeLModel=gemini-2.5-pro2026.051.40