Package software.amazon.awscdk.services.bedrockagentcore

Interface EvaluatorInferenceConfig

All Superinterfaces:

software.amazon.jsii.JsiiSerializable

All Known Implementing Classes:

EvaluatorInferenceConfig.Jsii$Proxy

@Generated(value="jsii-pacmak/1.130.0 (build 048a5ee)",
           date="2026-05-19T19:44:37.075Z")
@Stability(Stable)
public interface EvaluatorInferenceConfig
extends software.amazon.jsii.JsiiSerializable

Inference configuration for a custom LLM-as-a-Judge evaluator.

Controls how the foundation model generates evaluation responses.

Example:

 // LLM-as-a-Judge with categorical rating scale
 Evaluator categoricalEvaluator = Evaluator.Builder.create(this, "CategoricalEvaluator")
         .evaluatorName("domain_accuracy_evaluator")
         .level(EvaluationLevel.SESSION)
         .description("Evaluates domain-specific accuracy of agent responses")
         .evaluatorConfig(EvaluatorConfig.llmAsAJudge(LlmAsAJudgeOptions.builder()
                 .instructions("Evaluate whether the agent response is accurate within the healthcare domain.")
                 .modelId("us.anthropic.claude-sonnet-4-6")
                 .ratingScale(EvaluatorRatingScale.categorical(List.of(CategoricalRatingOption.builder().label("Accurate").definition("The response contains factually correct healthcare information.").build(), CategoricalRatingOption.builder().label("Inaccurate").definition("The response contains incorrect or misleading healthcare information.").build())))
                 .build()))
         .build();
 // LLM-as-a-Judge with numerical rating scale and inference config
 Evaluator numericalEvaluator = Evaluator.Builder.create(this, "NumericalEvaluator")
         .evaluatorName("response_quality_evaluator")
         .level(EvaluationLevel.TRACE)
         .evaluatorConfig(EvaluatorConfig.llmAsAJudge(LlmAsAJudgeOptions.builder()
                 .instructions("Rate the overall quality of the agent response on a scale of 1 to 5.")
                 .modelId("us.anthropic.claude-sonnet-4-6")
                 .ratingScale(EvaluatorRatingScale.numerical(List.of(NumericalRatingOption.builder().label("Poor").definition("Inadequate response.").value(1).build(), NumericalRatingOption.builder().label("Below Average").definition("Partially addresses the query.").value(2).build(), NumericalRatingOption.builder().label("Average").definition("Adequately addresses the query.").value(3).build(), NumericalRatingOption.builder().label("Good").definition("Well-structured and accurate response.").value(4).build(), NumericalRatingOption.builder().label("Excellent").definition("Outstanding response exceeding expectations.").value(5).build())))
                 .inferenceConfig(EvaluatorInferenceConfig.builder()
                         .maxTokens(1024)
                         .temperature(0.1)
                         .build())
                 .build()))
         .build();
 

Nested Class Summary

Nested Classes

Modifier and Type	Interface	Description
static final class	EvaluatorInferenceConfig.Builder	A builder for EvaluatorInferenceConfig
static final class	EvaluatorInferenceConfig.Jsii$Proxy	An implementation for EvaluatorInferenceConfig

Method Summary

Modifier and Type	Method	Description
static EvaluatorInferenceConfig.Builder	builder()
default Number	getMaxTokens()	The maximum number of tokens to generate in the model response.
default Number	getTemperature()	The temperature value that controls randomness in the model's responses.
default Number	getTopP()	The top-p sampling parameter that controls the diversity of the model's responses.

Methods inherited from interface software.amazon.jsii.JsiiSerializable

$jsii$toJson

Method Details

getMaxTokens

@Stability(Stable)
@Nullable
default Number getMaxTokens()

The maximum number of tokens to generate in the model response.

Default: - The foundation model's default maximum token limit is used

getTemperature

@Stability(Stable)
@Nullable
default Number getTemperature()

The temperature value that controls randomness in the model's responses.

Higher values produce more diverse outputs. Range: 0.0 to 1.0.

Default: - The foundation model's default temperature is used

getTopP

@Stability(Stable)
@Nullable
default Number getTopP()

The top-p sampling parameter that controls the diversity of the model's responses.

Range: 0.0 to 1.0.

Default: - The foundation model's default top-p value is used

builder

@Stability(Stable)
static EvaluatorInferenceConfig.Builder builder()

Returns:

a EvaluatorInferenceConfig.Builder of EvaluatorInferenceConfig