MetricsAll MetricsSingle-Turn

Contextual Relevancy

Contextual Relevancy is a single-turn metric used to evaluate a RAG retriever

Overview

The contextual relevancy metric is a single-turn RAG metric that uses LLM-as-a-judge to evaluate whether all retrieved context is relevant to the input query.

The input of a test case should not contain the entire prompt, but just the query when using the contextual relevancy metric.

Required Parameters

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

input
stringRequired

The input query you supply to your RAG application.

expected_output
stringRequired

The expected output your RAG application has to generate for a given input.

retrieval_context
list of stringRequired

The retrieved context your retriever outputs for a given input sorted by their rank.

How Is It Calculated?

The contextual relevancy metric first extracts independent statements from all retrieved context using an LLM, then uses the same LLM to determine how many of those statements are relevant to the input query.


Contextual Relevancy=Number of Relevant StatementsTotal Number of Statements\text{Contextual Relevancy} = \frac{\text{Number of Relevant Statements}}{\text{Total Number of Statements}}

The final score is the proportion of relevant statements in retrieval context.

Create Locally

You can create the ContextualRelevancyMetric in deepeval as follows:

1from deepeval.metrics import ContextualRelevancyMetric
2
3metric = ContextualRelevancyMetric()

Here’s a list of parameters you can configure when creating a ContextualRelevancyMetric:

threshold
numberDefaults to 0.5

A float to represent the minimum passing threshold.

model
string | ObjectDefaults to gpt-4.1

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

include_reason
booleanDefaults to true

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

async_mode
booleanDefaults to true

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

strict_mode
booleanDefaults to false

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

verbose_mode
booleanDefaults to false

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

evaluation_template
ContextualRelevancyTemplateDefaults to deepeval's template

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

This can be used for both single-turn E2E and component-level testing.

Create Remotely

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

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