Answer Relevancy

Answer relevancy is a single-turn metric to evaluate RAG generators

Overview

The answer relevancy metric uses LLM-as-a-judge to assess whether your RAG generator's output is relevant to the given input. It is a single-turn metric designed specifically for evaluating RAG QA specifically, and not general RAG.

Required Parameters

These are the parameters you must supply in your test case to run evaluations for answer relevancy metric:

inputstringrequired

The input query you supply to your RAG application.

actual_outputstringrequired

The final output your RAG application's generator generates.

How Is It Calculated?

The answer relevancy metric first breaks down the actual output of a test case into distinct statements, then calculates the proportion of those statements that are relevant to the given input.

\text{Answer Relevanacy} = \frac{\text{Number of Relevant Statements}}{\text{Total Number of Statements}}

The final score is the proportion of relevant statements found in the actual output.

Create Locally

You can create the AnswerRelevancyMetric in deepeval as follows:

from deepeval.metrics import AnswerRelevancyMetric

metric = AnswerRelevancyMetric()

Here's a list of parameters you can configure when creating a AnswerRelevancyMetric:

thresholdnumberdefault: 0.5

A float to represent the minimum passing threshold.

modelstring | Objectdefault: gpt-4.1

A string specifying which of OpenAI's GPT models to use OR any custom LLM model of type DeepEvalBaseLLM.

include_reasonbooleandefault: true

A boolean to enable the inclusion a reason for its evaluation score.

async_modebooleandefault: true

A boolean to enable concurrent execution within the measure() method.

strict_modebooleandefault: false

A boolean to enforce a binary metric score: 0 for perfection, 1 otherwise.

verbose_modebooleandefault: false

A boolean to print the intermediate steps used to calculate the metric score.

evaluation_templateAnswerRelevancyTemplatedefault: deepeval's template

An instance of AnswerRelevancyTemplate object, which allows you to override the default prompts used to compute the AnswerRelevancyMetric score.

Create Remotely

For users not using deepeval python, or want to run evals remotely on Confident AI, you can use the answer relevancy metric by adding it to a single-turn metric collection. This will allow you to use answer relevancy metric for:

Single-turn E2E testing
Single-turn component-level testing
Online and offline evals for traces and spans