> ## Documentation Index > Fetch the complete documentation index at: https://developers.phrase.com/llms.txt > Use this file to discover all available pages before exploring further. # Evaluate segments > Evaluates all segments against the AI Checks resolved from the given Content Group. When no checks are associated with the Content Group, all segments are returned with an empty `results` array — no error is raised. Requires **ADMIN** or **OWNER** IDM role. ## OpenAPI ````yaml /openapi/phrase-quality-evaluator.json post /v3/evaluation openapi: 3.1.2 info: title: Quality Evaluator API description: > The Quality Evaluator API enables you to evaluate the quality of translated segments using AI-powered checks. Create custom AI Checks with specific quality requirements, organize them into Quality Profiles, and run evaluations to identify translation issues automatically. version: 1.0.0 termsOfService: https://phrase.com/terms/ contact: name: Phrase Support url: https://support.phrase.com servers: - url: https://eu.phrase.com/quality-evaluator description: Production EU - url: https://us.phrase.com/quality-evaluator description: Production US security: - bearerAuth: [] tags: - name: AI Checks description: Create and manage AI Checks with custom quality requirements. - name: Analytics description: Retrieve aggregated evaluation analytics. - name: Evaluation description: Evaluate translation quality using AI Checks. - name: Quality Profiles description: Organize AI Checks into reusable Quality Profiles. paths: /v3/evaluation: post: tags: - Evaluation summary: Evaluate segments description: >- Evaluates all segments against the AI Checks resolved from the given Content Group. When no checks are associated with the Content Group, all segments are returned with an empty `results` array — no error is raised. Requires **ADMIN** or **OWNER** IDM role. operationId: evaluateV3 parameters: - name: Content-Type in: header required: true schema: type: string example: application/json requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/EvaluationV3Request' example: contentGroupId: cg-abc123 segments: - id: segment-1 source: Hello, world! target: Hallo, Welt! sourceLocaleCode: en_us targetLocaleCode: de_de responses: '200': description: >- Evaluation completed successfully. Returns evaluated segments with per-check results. content: application/json: schema: $ref: '#/components/schemas/EvaluationV3Response' example: contentGroupId: cg-abc123 segments: - id: segment-1 source: Hello, world! target: Hallo, Welt! results: - type: AI_CHECK ruleUid: rule-abc123 evaluation: isValid: true explanation: >- The translation uses standard German sentence structure. '400': description: >- Bad Request — missing required fields, a segment is missing `id`, or segment `id` values are not unique within the request. content: application/problem+json: schema: $ref: '#/components/schemas/ProblemDetail' '401': $ref: '#/components/responses/UnauthorizedResponse' '403': $ref: '#/components/responses/ForbiddenResponse' '503': description: Service Unavailable — evaluation was interrupted; retry the request. content: application/problem+json: schema: $ref: '#/components/schemas/ProblemDetail' default: $ref: '#/components/responses/ErrorResponse' components: schemas: EvaluationV3Request: type: object description: >- Request to evaluate translation segments using AI Checks resolved from a Content Group. properties: contentGroupId: type: string description: ID of the Content Group whose AI Checks are used for evaluation. segments: type: array description: >- Segments to evaluate. Each segment's `id` is required and must be unique within the request. minItems: 1 maxItems: 200 items: $ref: '#/components/schemas/RequestSegmentV3' sourceLocaleCode: $ref: '#/components/schemas/LocaleCode' targetLocaleCode: $ref: '#/components/schemas/LocaleCode' required: - contentGroupId - segments - sourceLocaleCode - targetLocaleCode EvaluationV3Response: type: object description: Evaluation results for the v3 endpoint. properties: contentGroupId: type: string description: ID of the Content Group whose AI Checks were used for evaluation. segments: type: array description: Evaluated segments with all check results nested per segment. items: $ref: '#/components/schemas/EvaluatedSegmentV3' required: - segments ProblemDetail: type: object description: Error response following RFC 9457. properties: type: type: string description: A URI reference identifying the problem type. example: urn:problem-type:unauthorized title: type: string description: A short, human-readable summary of the problem. example: Unauthorized status: type: integer description: The HTTP status code. example: 401 detail: type: string description: A human-readable explanation specific to this occurrence. example: Valid credentials were not supplied instance: type: string description: A URI reference identifying the specific occurrence. RequestSegmentV3: type: object description: >- A translation segment to evaluate. `id` is required and must be unique among all segments in the request; a missing or duplicate `id` results in a 400 Bad Request. properties: id: type: string description: >- Identifier for the segment. Required and must be unique within the request. source: type: string description: Source text that was translated. target: type: string description: Translated text to evaluate. required: - id - source - target LocaleCode: type: string description: >- Locale code compatible with Phrase locale format. Case-insensitive, supports both underscore and hyphen separators (e.g., `en_us`, `en-US`, `de_de`, `fr_fr`). example: en_us EvaluatedSegmentV3: type: object description: A single segment with all evaluation results from v3 checks. properties: id: type: string description: >- Segment identifier, matching the `id` supplied for this segment in the request. source: type: string description: Source text. target: type: string description: Translated text. results: type: array description: Results from all checks for this segment. items: $ref: '#/components/schemas/CheckResultV3' required: - id - source - target - results CheckResultV3: type: object description: Result from a single check for a segment. properties: type: $ref: '#/components/schemas/EvaluationItemType' description: Type of check. ruleUid: type: string description: UID of the rule that produced this result. evaluation: $ref: '#/components/schemas/CheckEvaluationV3' required: - type - evaluation EvaluationItemType: type: string description: Discriminator for the type of check that produced this result. enum: - AI_CHECK - QA_CHECK CheckEvaluationV3: type: object description: >- Evaluation result for a v3 check. `isValid` indicates whether the segment passed, `explanation` provides the reasoning. When `hasError` is true, the LLM call failed. properties: isValid: type: boolean description: Whether the segment passed this check. explanation: type: string description: Reasoning from the check explaining the evaluation result. hasError: type: boolean description: True when the check could not produce a result (LLM failure). errorMessage: type: string description: Error description when hasError is true. responses: UnauthorizedResponse: description: Authentication failed or credentials not provided. content: application/problem+json: schema: $ref: '#/components/schemas/ProblemDetail' example: type: urn:problem-type:unauthorized title: Unauthorized status: 401 detail: Valid credentials were not supplied ForbiddenResponse: description: >- Insufficient permissions. This operation requires the ADMIN or OWNER IDM role. content: application/problem+json: schema: $ref: '#/components/schemas/ProblemDetail' example: type: urn:problem-type:forbidden title: Forbidden status: 403 detail: Insufficient permissions to perform this action ErrorResponse: description: An error occurred. content: application/problem+json: schema: $ref: '#/components/schemas/ProblemDetail' securitySchemes: bearerAuth: type: http scheme: bearer description: >- Use a Bearer token obtained from Phrase Platform authentication. See the Authentication guide for details on how to obtain a token. ````