openapi: 3.0.3
info:
  title: Data Plane
  version: '2.0'
  contact:
    name: Seldon Technologies Ltd.
    url: 'https://www.seldon.io/'
    email: hello@seldon.io
  description: REST protocol to interact with inference servers.
  license:
    name: Business Source License 1.1
servers: []
paths:
  /v2/health/live:
    get:
      summary: Server Live
      responses:
        '200':
          description: OK
      operationId: server-live
      description: The “server live” API indicates if the inference server is able to receive and respond to metadata and inference requests. The “server live” API can be used directly to implement the Kubernetes `livenessProbe`.
      tags:
        - health
        - server
  /v2/health/ready:
    get:
      summary: Server Ready
      tags:
        - health
        - server
      responses:
        '200':
          description: OK
      operationId: server-ready
      description: The “server ready” health API indicates if all the models are ready for inferencing. The “server ready” health API can be used directly to implement the Kubernetes readinessProbe.
  '/v2/models/{model_name}/versions/{model_version}/ready':
    parameters:
      - schema:
          type: string
        name: model_name
        in: path
        required: true
      - schema:
          type: string
        name: model_version
        in: path
        required: true
    get:
      summary: Model Ready
      tags:
        - health
        - model
      responses:
        '200':
          description: OK
      operationId: model-version-ready
      description: The “model ready” health API indicates if a specific model is ready for inferencing. The model name and (optionally) version must be available in the URL. If a version is not provided the server may choose a version based on its own policies.
  '/v2/models/{model_name}/ready':
    parameters:
      - schema:
          type: string
        name: model_name
        in: path
        required: true
    get:
      summary: Model Ready
      tags:
        - health
        - model
      responses:
        '200':
          description: OK
      operationId: model-ready
      description: The “model ready” health API indicates if a specific model is ready for inferencing. The model name and (optionally) version must be available in the URL. If a version is not provided the server may choose a version based on its own policies.
  /v2:
    get:
      summary: Server Metadata
      tags:
        - metadata
        - server
      responses:
        '200':
          description: OK
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/MetadataServerResponse'
        '400':
          description: Bad Request
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/MetadataServerErrorResponse'
      operationId: server-metadata
      description: The server metadata endpoint provides information about the server. A server metadata request is made with an HTTP GET to a server metadata endpoint.
  '/v2/models/{model_name}/versions/{model_version}':
    parameters:
      - schema:
          type: string
        name: model_name
        in: path
        required: true
      - schema:
          type: string
        name: model_version
        in: path
        required: true
    get:
      summary: Model Metadata
      tags:
        - metadata
        - model
      responses:
        '200':
          description: OK
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/MetadataModelResponse'
        '404':
          description: Not Found
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/MetadataModelErrorResponse'
      operationId: model-version-metadata
      description: The per-model metadata endpoint provides information about a model. A model metadata request is made with an HTTP GET to a model metadata endpoint. The model name and (optionally) version must be available in the URL. If a version is not provided the server may choose a version based on its own policies or return an error.
  '/v2/models/{model_name}':
    parameters:
      - schema:
          type: string
        name: model_name
        in: path
        required: true
    get:
      summary: Model Metadata
      tags:
        - metadata
        - model
      responses:
        '200':
          description: OK
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/MetadataModelResponse'
        '404':
          description: Not Found
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/MetadataModelErrorResponse'
      operationId: model-metadata
      description: The per-model metadata endpoint provides information about a model. A model metadata request is made with an HTTP GET to a model metadata endpoint. The model name and (optionally) version must be available in the URL. If a version is not provided the server may choose a version based on its own policies or return an error.
  '/v2/models/{model_name}/versions/{model_version}/infer':
    parameters:
      - schema:
          type: string
        name: model_name
        in: path
        required: true
      - schema:
          type: string
        name: model_version
        in: path
        required: true
    post:
      summary: Model Inference
      operationId: model-version-inference
      responses:
        '200':
          description: OK
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/InferenceResponse'
        '400':
          description: Bad Request
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/InferenceErrorResponse'
        '404':
          description: Not Found
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/InferenceErrorResponse'
        '500':
          description: Internal Server Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/InferenceErrorResponse'
      requestBody:
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/InferenceRequest'
      description: An inference request is made with an HTTP POST to an inference endpoint. The model name and (optionally) version must be available in the URL. If a version is not provided the server may choose a version based on its own policies or return an error.
      tags:
        - inference
        - model
  '/v2/models/{model_name}/infer':
    parameters:
      - schema:
          type: string
        name: model_name
        in: path
        required: true
    post:
      summary: Model Inference
      operationId: model-inference
      responses:
        '200':
          description: OK
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/InferenceResponse'
        '400':
          description: Bad Request
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/InferenceErrorResponse'
        '404':
          description: Not Found
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/InferenceErrorResponse'
        '500':
          description: Internal Server Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/InferenceErrorResponse'
      requestBody:
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/InferenceRequest'
      description: An inference request is made with an HTTP POST to an inference endpoint. The model name and (optionally) version must be available in the URL. If a version is not provided the server may choose a version based on its own policies or return an error.
      tags:
        - inference
        - model
components:
  schemas:
    MetadataServerResponse:
      title: MetadataServerResponse
      type: object
      description: ''
      x-examples: {}
      properties:
        name:
          type: string
        version:
          type: string
        extensions:
          type: array
          items:
            type: string
      required:
        - name
        - version
        - extensions
    MetadataModelResponse:
      title: MetadataModelResponse
      type: object
      properties:
        name:
          type: string
        versions:
          type: array
          items:
            type: string
        platform:
          type: string
        inputs:
          type: array
          items:
            $ref: '#/components/schemas/MetadataTensor'
        outputs:
          type: array
          items:
            $ref: '#/components/schemas/MetadataTensor'
        parameters:
          $ref: '#/components/schemas/Parameters'
      required:
        - name
        - platform
    MetadataServerErrorResponse:
      title: MetadataServerErrorResponse
      type: object
      properties:
        error:
          type: string
      required:
        - error
    MetadataTensor:
      title: MetadataTensor
      type: object
      properties:
        name:
          type: string
        datatype:
          type: string
        shape:
          type: array
          items:
            type: integer
        parameters:
          $ref: '#/components/schemas/Parameters'
      required:
        - name
        - datatype
        - shape
    MetadataModelErrorResponse:
      title: MetadataModelErrorResponse
      type: object
      properties:
        error:
          type: string
      required:
        - error
    InferenceRequest:
      title: InferenceRequest
      type: object
      x-examples:
        Example 1:
          id: '42'
          inputs:
            - name: input0
              shape:
                - 2
                - 2
              datatype: UINT32
              data:
                - 1
                - 2
                - 3
                - 4
            - name: input1
              shape:
                - 3
              datatype: BOOL
              data:
                - true
          outputs:
            - name: output0
        Example 2:
          id: '42'
          outputs:
            - name: output0
              shape:
                - 3
                - 2
              datatype: FP32
              data:
                - 1
                - 1.1
                - 2
                - 2.1
                - 3
                - 3.1
      properties:
        id:
          type: string
        parameters:
          $ref: '#/components/schemas/Parameters'
        inputs:
          type: array
          items:
            $ref: '#/components/schemas/RequestInput'
        outputs:
          type: array
          items:
            $ref: '#/components/schemas/RequestOutput'
      required:
        - inputs
    Parameters:
      type: object
      title: Parameters
      x-examples: {}
      additionalProperties: true
      properties:
        content_type:
          type: string
        headers:
          type: object
    RequestInput:
      title: RequestInput
      type: object
      properties:
        name:
          type: string
        shape:
          type: array
          items:
            type: integer
        datatype:
          type: string
        parameters:
          $ref: '#/components/schemas/Parameters'
        data:
          $ref: '#/components/schemas/TensorData'
      required:
        - name
        - shape
        - datatype
        - data
    TensorData:
      title: TensorData
    RequestOutput:
      title: RequestOutput
      type: object
      properties:
        name:
          type: string
        parameters:
          $ref: '#/components/schemas/Parameters'
      required:
        - name
    ResponseOutput:
      title: ResponseOutput
      type: object
      properties:
        name:
          type: string
        shape:
          type: array
          items:
            type: integer
        datatype:
          type: string
        parameters:
          $ref: '#/components/schemas/Parameters'
        data:
          $ref: '#/components/schemas/TensorData'
      required:
        - name
        - shape
        - datatype
        - data
    InferenceResponse:
      title: InferenceResponse
      type: object
      properties:
        model_name:
          type: string
        model_version:
          type: string
        id:
          type: string
        parameters:
          $ref: '#/components/schemas/Parameters'
        outputs:
          type: array
          items:
            $ref: '#/components/schemas/ResponseOutput'
      required:
        - model_name
        - outputs
    InferenceErrorResponse:
      title: InferenceErrorResponse
      type: object
      properties:
        error:
          type: string
tags:
  - name: health
  - name: inference
  - name: metadata
  - name: model
  - name: server