> ## Documentation Index
> Fetch the complete documentation index at: https://penseapp.vercel.app/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Create evaluator version

> Add a new version to an evaluator you created



## OpenAPI

````yaml /api-reference/openapi.json post /evaluators/{evaluator_uuid}/versions
openapi: 3.1.0
info:
  title: Calibrate Public API
  version: 0.1.0
  description: Programmatic API for CI/automation. Pass your key in the `X-API-Key` header.
servers:
  - url: https://api.calibrate.artpark.ai
    description: Production
security: []
paths:
  /evaluators/{evaluator_uuid}/versions:
    post:
      tags:
        - evaluators
      summary: Create evaluator version
      description: Add a new version to an evaluator you created
      operationId: create_version_evaluators__evaluator_uuid__versions_post
      parameters:
        - name: evaluator_uuid
          in: path
          required: true
          schema:
            type: string
            description: Evaluator to add a version to
            examples:
              - f47ac10b-58cc-4372-a567-0e02b2c3d479
            title: Evaluator Uuid
          description: Evaluator to add a version to
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/EvaluatorVersionCreateRequest'
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/VersionCreateResponse'
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
      security:
        - ApiKeyAuth: []
components:
  schemas:
    EvaluatorVersionCreateRequest:
      properties:
        judge_model:
          type: string
          title: Judge Model
          description: >-
            The model that runs the judge, named the way its provider does, for
            example `openai/gpt-4.1` or `anthropic/claude-sonnet-4`
        system_prompt:
          type: string
          title: System Prompt
          description: Judge system prompt. May contain `{{variable}}` placeholders
        output_config:
          anyOf:
            - $ref: '#/components/schemas/OutputConfig'
            - type: 'null'
          description: >-
            The scale points and their labels. Required for a `rating`
            evaluator. A `binary` evaluator uses the default Correct/Wrong
            labels unless you set your own
        variables:
          anyOf:
            - items:
                $ref: '#/components/schemas/VariableSpec'
              type: array
            - type: 'null'
          title: Variables
          description: >-
            Declared prompt variables. Omit if the prompt has none. After the
            first version the variable names are fixed. You can change a
            variable's description or default, but not add, remove, or rename
            one
        make_live:
          type: boolean
          title: Make Live
          description: >-
            When `true`, immediately point the evaluator's live version at this
            new version
          default: false
      type: object
      required:
        - judge_model
        - system_prompt
      title: EvaluatorVersionCreateRequest
      description: Request body for adding a version to an existing evaluator.
    VersionCreateResponse:
      properties:
        version_uuid:
          type: string
          maxLength: 36
          minLength: 36
          title: Version Uuid
          description: ID of the newly created version
          examples:
            - 6ba7b810-9dad-11d1-80b4-00c04fd430c8
        version_number:
          type: integer
          title: Version Number
          description: >-
            The version's number. The first version is 1, and it goes up by one
            for each new version of the evaluator
      type: object
      required:
        - version_uuid
        - version_number
      title: VersionCreateResponse
    HTTPValidationError:
      properties:
        detail:
          items:
            $ref: '#/components/schemas/ValidationError'
          type: array
          title: Detail
      type: object
      title: HTTPValidationError
    OutputConfig:
      properties:
        scale:
          anyOf:
            - items:
                $ref: '#/components/schemas/OutputScaleEntry'
              type: array
            - type: 'null'
          title: Scale
          description: >-
            The ordered scale points that make up the rubric, each with its
            label
      type: object
      title: OutputConfig
    VariableSpec:
      properties:
        name:
          type: string
          title: Name
          description: Name of a `{{placeholder}}` used in the system prompt
        description:
          anyOf:
            - type: string
            - type: 'null'
          title: Description
          description: What the variable is for. Omit if self-evident
        default:
          anyOf:
            - type: string
            - type: 'null'
          title: Default
          description: Default value used when you omit this variable. Omit for no default
      type: object
      required:
        - name
      title: VariableSpec
    ValidationError:
      properties:
        loc:
          items:
            anyOf:
              - type: string
              - type: integer
          type: array
          title: Location
        msg:
          type: string
          title: Message
        type:
          type: string
          title: Error Type
      type: object
      required:
        - loc
        - msg
        - type
      title: ValidationError
    OutputScaleEntry:
      properties:
        value:
          title: Value
          description: >-
            The value for this scale point. Use a boolean for a `binary`
            evaluator, a number for a `rating` one
        name:
          type: string
          title: Name
          description: Short label for this scale point
        description:
          anyOf:
            - type: string
            - type: 'null'
          title: Description
          description: >-
            Rubric text for this level, added to the judge prompt. Omit to leave
            this level undescribed
        color:
          anyOf:
            - type: string
            - type: 'null'
          title: Color
          description: Color to show for this level. Omit for the default
      type: object
      required:
        - value
        - name
      title: OutputScaleEntry
      description: >-
        One entry in an evaluator's output_config.scale. `value` is
        bool|number|string

        depending on output_type.
  securitySchemes:
    ApiKeyAuth:
      type: apiKey
      in: header
      name: X-API-Key
      description: API key for authentication

````