II.
Benchmark overview
Reference · livebenchmark:agentclinic
AgentClinic overview
Multimodal medical-agent benchmark simulating a clinical encounter (patient, exam, lab tools, instruments) to evaluate diagnostic reasoning end-to-end.
Attributes
displayName
AgentClinic
homepageUrl
kind
domain-specific
targetsKind
AgentVersion
description
Multimodal medical-agent benchmark simulating a clinical encounter
(patient, exam, lab tools, instruments) to evaluate diagnostic
reasoning end-to-end.
Outgoing edges
applies_to1
- domain:healthcare·DomainHealthcare
covers1
- skill-area:medical-agent·SkillAreaMedical Agent
Incoming edges
belongs_to_benchmark1
- test-set:agentclinic-clinical-cases·TestSetAgentClinic clinical case suite