Metrics

Overview

Heuristic metrics

Hallucination

LLM Juries

G-Eval

Conversation-level GEval Metrics

Compliance risk

Prompt uncertainty

Moderation

Meaning Match

Usefulness

Summarization consistency

Summarization coherence

Dialogue helpfulness

Answer relevance

Context precision

Context recall

Trajectory accuracy

Agent task completion

Agent tool correctness

Conversational metrics

Custom model

Advanced configuration

Custom metric

Custom conversation metric

Structured Output Compliance

Task span metrics