Function createRetrievalRelevanceEvaluator

createRetrievalRelevanceEvaluator<
    RecordType extends
        Record<string, unknown> = RetrievalRelevanceEvaluationRecord,
>(
    args: RetrievalRelevanceEvaluatorArgs<RecordType>,
): ClassificationEvaluator<RecordType>
Creates a retrieval relevance evaluator function.

This function returns an evaluator that determines whether the external information retrieved during a step is relevant to the request it was serving. Unlike createDocumentRelevanceEvaluator, which judges a single document against a question, this evaluator is source-agnostic and scores the retrieved information as a whole (holistically, per retrieval step), whether it came from a vector search, a tool or MCP call, a web search, or content embedded in an LLM turn.
Type Parameters
- RecordType extends Record<string, unknown> = RetrievalRelevanceEvaluationRecord
Parameters
- args: RetrievalRelevanceEvaluatorArgs<RecordType>
  The arguments for creating the retrieval relevance evaluator.
  - Optionalchoices?: ClassificationChoicesMap
  - OptionalinputMapping?: ObjectMapping<RecordType>
    
    The mapping of the input to evaluate to the shape that the evaluator expects
  - model: LanguageModel
  - Optionalname?: string
  - OptionaloptimizationDirection?: OptimizationDirection
  - OptionalpromptTemplate?: PromptTemplate
  - Optionaltelemetry?: TelemetryConfig
Returns ClassificationEvaluator<RecordType>
An evaluator function that takes a RetrievalRelevanceEvaluationRecord and returns a classification result indicating whether the retrieved information is relevant or irrelevant to the request.
Example
```
const evaluator = createRetrievalRelevanceEvaluator({ model: openai("gpt-4o-mini") });
const result = await evaluator.evaluate({
  input: "What is the capital of France?",
  context: "Paris is the capital and largest city of France.",
});
console.log(result.label); // "relevant"
```
- Defined in llm/createRetrievalRelevanceEvaluator.ts:65