@AI Evy on the retell tests endpoint there is a parameter called metrics for evaluation. what is expected to be passed there?

Thanks for tagging me @Nicolas. I'll look through relevant sources and get back to you shortly.

@AI Evy i mean here: https://docs.retellai.com/api-references/create-test-case-definition there is a metrics, but not sure what we could pass there.

### Answer Based on the AI QA Metrics documentation, the metrics parameter in the test case definition expects specific quality assurance metrics that evaluate your AI agent's performance. [(1) ]() The available metrics you can pass include: Speech Quality Metrics: - Natural Tonality Rate - evaluates if agent speech sounds natural and human-like AI Accuracy Metrics: - LLM Hallucination Rate - measures how often the LLM generates incorrect or fabricated information - Agent Hallucination - tracks how often the agent hallucinates during conversations Knowledge Base Metrics: - KB Recall - measures how effectively the agent retrieved and used relevant information from the knowledge base Tool and Function Metrics: - Tool Call Accuracy - measures the rate at which the agent correctly invoked tools or functions - Tool Call Inaccuracy - measures the rate of incorrect tool invocations - Custom Tool Success Rate - percentage of custom tool calls that completed successfully - Avg Custom Tool Latency - average time for custom tools to execute and return results Conversation Flow Metrics: - Transition Accuracy - measures accuracy of transitions between conversation nodes or states - Node Transition Inaccuracy - measures incorrect node transitions in conversation flows Call Resolution Metrics: - Call Resolution Rate - percentage of calls successfully resolved according to your defined criteria - Average Score - overall quality score based on resolution criteria and weighted scoring configuration These metrics help you evaluate agent performance using defined thresholds and success criteria during simulation and batch testing. [(2) ]() [(1) ]()