Aleph Alpha Benchmark Executions API

openapi: 3.2.0
info:
  title: PhariaStudio Benchmark Executions API
  version: 0.1.0
  description: Endpoints for handling execution instances of benchmarks. These provides an overview of all evaluation steps during execution (run, evaluation, aggregation).
servers:
- url: '{host}/v1/studio'
  variables:
    host:
      default: https://api.pharia.example.com
security:
- BearerAuth: []
tags:
- name: Benchmark Executions
  description: Endpoints for handling execution instances of benchmarks. These provides an overview of all evaluation steps during execution (run, evaluation, aggregation).
paths:
  /projects/{project_id}/evaluation/benchmarks/{benchmark_id}/executions:
    post:
      tags:
      - Benchmark Executions
      summary: Create Benchmark Execution
      description: Create a new `BenchmarkExecution`.
      operationId: create_benchmark_execution_projects__project_id__evaluation_benchmarks__benchmark_id__executions_post
      parameters:
      - name: project_id
        in: path
        required: true
        schema:
          type: string
          format: uuid
          title: Project Id
      - name: benchmark_id
        in: path
        required: true
        schema:
          type: string
          format: uuid
          title: Benchmark Id
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/PostBenchmarkExecutionRequest'
      responses:
        '201':
          description: Successful Response
          content:
            application/json:
              schema:
                type: string
                format: uuid
                title: Response Create Benchmark Execution Projects  Project Id  Evaluation Benchmarks  Benchmark Id  Executions Post
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
    get:
      tags:
      - Benchmark Executions
      summary: Get Many Benchmark Executions
      description: Get multiple `BenchmarkExecution`s. Results are sorted by descending creation date.
      operationId: get_many_benchmark_executions_projects__project_id__evaluation_benchmarks__benchmark_id__executions_get
      parameters:
      - name: project_id
        in: path
        required: true
        schema:
          type: string
          format: uuid
          title: Project Id
      - name: benchmark_id
        in: path
        required: true
        schema:
          type: string
          format: uuid
          title: Benchmark Id
      - name: search
        in: query
        required: false
        schema:
          anyOf:
          - type: string
            minLength: 1
          - type: 'null'
          description: Search term to find results containing this text in any searchable field (case insensitive).
          title: Search
        description: Search term to find results containing this text in any searchable field (case insensitive).
      - name: sort_by
        in: query
        required: false
        schema:
          anyOf:
          - $ref: '#/components/schemas/BenchmarkExecutionSortableColumns'
          - type: string
          description: The column to sort by. Can be a predefined column or 'statistics.KEY' for JSONB sorting.
          default: created_at
          title: Sort By
        description: The column to sort by. Can be a predefined column or 'statistics.KEY' for JSONB sorting.
      - name: sort_direction
        in: query
        required: false
        schema:
          $ref: '#/components/schemas/SortDirection'
          description: The direction to sort by.
          default: desc
        description: The direction to sort by.
      - name: page
        in: query
        required: false
        schema:
          type: integer
          exclusiveMinimum: 0
          description: The page number to retrieve. Starts at 1.
          default: 1
          title: Page
        description: The page number to retrieve. Starts at 1.
      - name: size
        in: query
        required: false
        schema:
          type: integer
          maximum: 100
          exclusiveMinimum: 0
          description: The number of items per page.
          default: 25
          title: Size
        description: The number of items per page.
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/PagedResponse_BenchmarkExecutionResponse_'
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
  /projects/{project_id}/evaluation/benchmarks/{benchmark_id}/executions/{execution_id}:
    get:
      tags:
      - Benchmark Executions
      summary: Get Benchmark Execution
      description: Get a `BenchmarkExecution` by ID. Gives an error if the project or benchmark do not exist and `None` if they do but `BenchmarkExecution` was not found.
      operationId: get_benchmark_execution_projects__project_id__evaluation_benchmarks__benchmark_id__executions__execution_id__get
      parameters:
      - name: project_id
        in: path
        required: true
        schema:
          type: string
          format: uuid
          title: Project Id
      - name: benchmark_id
        in: path
        required: true
        schema:
          type: string
          format: uuid
          title: Benchmark Id
      - name: execution_id
        in: path
        required: true
        schema:
          type: string
          format: uuid
          title: Execution Id
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema:
                anyOf:
                - $ref: '#/components/schemas/BenchmarkExecutionResponse'
                - type: 'null'
                title: Response Get Benchmark Execution Projects  Project Id  Evaluation Benchmarks  Benchmark Id  Executions  Execution Id  Get
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
    delete:
      tags:
      - Benchmark Executions
      summary: Delete Benchmark Execution
      description: Delete a `BenchmarkExecution`.
      operationId: delete_benchmark_execution_projects__project_id__evaluation_benchmarks__benchmark_id__executions__execution_id__delete
      parameters:
      - name: project_id
        in: path
        required: true
        schema:
          type: string
          format: uuid
          title: Project Id
      - name: benchmark_id
        in: path
        required: true
        schema:
          type: string
          format: uuid
          title: Benchmark Id
      - name: execution_id
        in: path
        required: true
        schema:
          type: string
          format: uuid
          title: Execution Id
      responses:
        '204':
          description: Successful Response
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
components:
  schemas:
    ValidationError:
      properties:
        loc:
          items:
            anyOf:
            - type: string
            - type: integer
          type: array
          title: Location
        msg:
          type: string
          title: Message
        type:
          type: string
          title: Error Type
        input:
          title: Input
        ctx:
          type: object
          title: Context
      type: object
      required:
      - loc
      - msg
      - type
      title: ValidationError
    BenchmarkExecutionSortableColumns:
      type: string
      enum:
      - name
      - description
      - eval_successful_count
      - run_success_avg_latency
      - run_success_avg_token_count
      - created_at
      - start
      - end
      - run_start
      - run_end
      - eval_start
      - eval_end
      - aggregation_start
      - aggregation_end
      - updated_at
      - created_by
      - updated_by
      title: BenchmarkExecutionSortableColumns
    HTTPValidationError:
      properties:
        detail:
          items:
            $ref: '#/components/schemas/ValidationError'
          type: array
          title: Detail
      type: object
      title: HTTPValidationError
    PostBenchmarkExecutionRequest:
      properties:
        name:
          type: string
          title: Name
          description: The name of the benchmark execution. This field is searchable.
          examples:
          - Benchmark Execution
        description:
          anyOf:
          - type: string
          - type: 'null'
          title: Description
          description: The description of the benchmark execution. This field is searchable.
          examples:
          - This is a benchmark execution.
        labels:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          title: Labels
        metadata:
          anyOf:
          - additionalProperties: true
            type: object
          - type: 'null'
          title: Metadata
        start:
          type: string
          format: date-time
          title: Start
          description: Start of the benchmark execution in UTC.
        end:
          type: string
          format: date-time
          title: End
          description: End of the benchmark execution in UTC. Does not include upload time.
        run_start:
          type: string
          format: date-time
          title: Run Start
          description: Start of running the given task on the benchmark dataset in UTC.
        run_end:
          type: string
          format: date-time
          title: Run End
          description: End of running the given task on the benchmark dataset in UTC.
        run_successful_count:
          type: integer
          title: Run Successful Count
          description: Number of examples that the task was successfully executed on.
        run_failed_count:
          type: integer
          title: Run Failed Count
          description: Number of examples that the task raised an exception on. These will be skipped in the evaluation.
        run_success_avg_latency:
          type: number
          title: Run Success Avg Latency
          description: Average time it takes the task to generate an output in ns. Does not include failed runs.
        run_success_avg_token_count:
          type: number
          title: Run Success Avg Token Count
          description: Average amount of tokens it takes the task to generate an output. Includes tokens from intermediate steps. Does not include failed runs.
        eval_start:
          type: string
          format: date-time
          title: Eval Start
          description: Start of evaluation process of the given task on the benchmark dataset in UTC.
        eval_end:
          type: string
          format: date-time
          title: Eval End
          description: End of evaluation process of the given task on the benchmark dataset in UTC.
        eval_successful_count:
          type: integer
          title: Eval Successful Count
          description: Number of evaluations of task runs that succeeded. Failed runs are skipped and not included.
        eval_failed_count:
          type: integer
          title: Eval Failed Count
          description: Number of evaluations of task runs that failed. Failed runs are skipped and not included.
        aggregation_start:
          type: string
          format: date-time
          title: Aggregation Start
          description: Start of the aggregation process in UTC.
        aggregation_end:
          type: string
          format: date-time
          title: Aggregation End
          description: End of the aggregation process in UTC.
        statistics:
          $ref: '#/components/schemas/JsonSerializable-Input'
          description: Data that is the result of the aggregation process.
      type: object
      required:
      - name
      - start
      - end
      - run_start
      - run_end
      - run_successful_count
      - run_failed_count
      - run_success_avg_latency
      - run_success_avg_token_count
      - eval_start
      - eval_end
      - eval_successful_count
      - eval_failed_count
      - aggregation_start
      - aggregation_end
      - statistics
      title: PostBenchmarkExecutionRequest
    SortDirection:
      type: string
      enum:
      - asc
      - desc
      title: SortDirection
    BenchmarkExecutionResponse:
      properties:
        name:
          type: string
          title: Name
          description: The name of the benchmark execution. This field is searchable.
          examples:
          - Benchmark Execution
        description:
          anyOf:
          - type: string
          - type: 'null'
          title: Description
          description: The description of the benchmark execution. This field is searchable.
          examples:
          - This is a benchmark execution.
        labels:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          title: Labels
        metadata:
          anyOf:
          - additionalProperties: true
            type: object
          - type: 'null'
          title: Metadata
        start:
          type: string
          format: date-time
          title: Start
          description: Start of the benchmark execution in UTC.
        end:
          type: string
          format: date-time
          title: End
          description: End of the benchmark execution in UTC. Does not include upload time.
        run_start:
          type: string
          format: date-time
          title: Run Start
          description: Start of running the given task on the benchmark dataset in UTC.
        run_end:
          type: string
          format: date-time
          title: Run End
          description: End of running the given task on the benchmark dataset in UTC.
        run_successful_count:
          type: integer
          title: Run Successful Count
          description: Number of examples that the task was successfully executed on.
        run_failed_count:
          type: integer
          title: Run Failed Count
          description: Number of examples that the task raised an exception on. These will be skipped in the evaluation.
        run_success_avg_latency:
          type: number
          title: Run Success Avg Latency
          description: Average time it takes the task to generate an output in ns. Does not include failed runs.
        run_success_avg_token_count:
          type: number
          title: Run Success Avg Token Count
          description: Average amount of tokens it takes the task to generate an output. Includes tokens from intermediate steps. Does not include failed runs.
        eval_start:
          type: string
          format: date-time
          title: Eval Start
          description: Start of evaluation process of the given task on the benchmark dataset in UTC.
        eval_end:
          type: string
          format: date-time
          title: Eval End
          description: End of evaluation process of the given task on the benchmark dataset in UTC.
        eval_successful_count:
          type: integer
          title: Eval Successful Count
          description: Number of evaluations of task runs that succeeded. Failed runs are skipped and not included.
        eval_failed_count:
          type: integer
          title: Eval Failed Count
          description: Number of evaluations of task runs that failed. Failed runs are skipped and not included.
        aggregation_start:
          type: string
          format: date-time
          title: Aggregation Start
          description: Start of the aggregation process in UTC.
        aggregation_end:
          type: string
          format: date-time
          title: Aggregation End
          description: End of the aggregation process in UTC.
        statistics:
          $ref: '#/components/schemas/JsonSerializable-Output'
          description: Data that is the result of the aggregation process.
        updated_at:
          type: string
          format: date-time
          title: Updated At
          description: The timestamp when this resource was last updated in UTC.
        updated_by:
          type: string
          title: Updated By
          description: The ID of the user that updated the resource.
          examples:
          - '12345'
        created_at:
          type: string
          format: date-time
          title: Created At
          description: The timestamp when this resource was created in UTC.
        created_by:
          type: string
          title: Created By
          description: The ID of the user that created the resource.
          examples:
          - '12345'
        id:
          type: string
          title: Id
        benchmark_id:
          type: string
          title: Benchmark Id
      type: object
      required:
      - name
      - start
      - end
      - run_start
      - run_end
      - run_successful_count
      - run_failed_count
      - run_success_avg_latency
      - run_success_avg_token_count
      - eval_start
      - eval_end
      - eval_successful_count
      - eval_failed_count
      - aggregation_start
      - aggregation_end
      - statistics
      - updated_at
      - updated_by
      - created_at
      - created_by
      - id
      - benchmark_id
      title: BenchmarkExecutionResponse
    JsonSerializable-Output:
      anyOf:
      - type: integer
      - type: number
      - type: string
      - type: boolean
      - items:
          $ref: '#/components/schemas/JsonSerializable-Output'
        type: array
      - additionalProperties:
          $ref: '#/components/schemas/JsonSerializable-Output'
        type: object
      - type: 'null'
    PagedResponse_BenchmarkExecutionResponse_:
      properties:
        total:
          type: integer
          title: Total
        page:
          type: integer
          title: Page
        size:
          type: integer
          title: Size
        num_pages:
          type: integer
          title: Num Pages
        items:
          items:
            $ref: '#/components/schemas/BenchmarkExecutionResponse'
          type: array
          title: Items
          default: []
      type: object
      required:
      - total
      - page
      - size
      - num_pages
      title: PagedResponse[BenchmarkExecutionResponse]
    JsonSerializable-Input:
      anyOf:
      - type: integer
      - type: number
      - type: string
      - type: boolean
      - items:
          $ref: '#/components/schemas/JsonSerializable-Input'
        type: array
      - additionalProperties:
          $ref: '#/components/schemas/JsonSerializable-Input'
        type: object
      - type: 'null'
  securitySchemes:
    BearerAuth:
      type: http
      scheme: bearer
      bearerFormat: JWT