# ModelJudgeEvaluator An evaluator that uses a language model as a judge to score responses. ``` struct ModelJudgeEvaluator where Input : ModelSampleProtocol ``` ## Overview `ModelJudgeEvaluator` sends the query, response, and optional reference data to a judge model, which returns scores for one or more dimensions. The response is automatically serialized as JSON, because `OutputType` is `Codable`, or is customizable via [`ModelJudgePrompt`](/documentation/Evaluations/ModelJudgePrompt). ## Topics ### Creating a single-dimension evaluator [`init(_:scale:judge:scoringMode:)`](/documentation/Evaluations/ModelJudgeEvaluator/init(_:scale:judge:scoringMode:)) Creates a single-metric evaluator with default prompt configuration. [`init(_:scale:judge:scoringMode:prompt:)`](/documentation/Evaluations/ModelJudgeEvaluator/init(_:scale:judge:scoringMode:prompt:)) Creates a single-metric evaluator with a custom judge prompt. ### Creating a multi-dimension evaluator [`init(judge:dimensions:scoringMode:)`](/documentation/Evaluations/ModelJudgeEvaluator/init(judge:dimensions:scoringMode:)) Creates a multi-metric evaluator with default prompt configuration. [`init(judge:dimensions:scoringMode:prompt:)`](/documentation/Evaluations/ModelJudgeEvaluator/init(judge:dimensions:scoringMode:prompt:)) Creates a multi-metric evaluator with a custom judge prompt. ### Creating a pairwise evaluator [`pairwise(_:scale:judge:scoringMode:evaluationTarget:)`](/documentation/Evaluations/ModelJudgeEvaluator/pairwise(_:scale:judge:scoringMode:evaluationTarget:)) Creates a pairwise comparison evaluator that compares the model’s response against the sample’s expected value. [`pairwise(judge:dimensions:scoringMode:evaluationTarget:)`](/documentation/Evaluations/ModelJudgeEvaluator/pairwise(judge:dimensions:scoringMode:evaluationTarget:)) Creates a multi-metric pairwise comparison evaluator. ### Configuring the judge prompt [`defaultInstructions`](/documentation/Evaluations/ModelJudgeEvaluator/defaultInstructions) The default system instructions the model uses when no custom instructions are provided. [`judgePrompt(for:output:)`](/documentation/Evaluations/ModelJudgeEvaluator/judgePrompt(for:output:)) Builds and returns the full judge prompt for inspection, debugging, or logging. ### Inspecting the evaluator [`dimensions`](/documentation/Evaluations/ModelJudgeEvaluator/dimensions) The dimensions this evaluator scores. [`scoringMode`](/documentation/Evaluations/ModelJudgeEvaluator/scoringMode) The scoring constraint mode. See [`ScoringMode`](/documentation/Evaluations/ScoringMode). [`ScoringMode`](/documentation/Evaluations/ScoringMode) The scoring constraint mode for a model-as-judge evaluator. ### Errors [`ModelJudgeError`](/documentation/Evaluations/ModelJudgeError) --- Copyright © 2026 Apple Inc. All rights reserved. | [Terms of Use](https://www.apple.com/legal/internet-services/terms/site.html) | [Privacy Policy](https://www.apple.com/privacy/privacy-policy)