Skip to main content
Version: 4.7.34

Create a new test result from runs​

Create a new test result from runs

Request Body required
    pipelineSlug string

    The pipeline slug to create the test result for. Only one of pipelineSlug or pipelineId is required.

    pipelineId string

    The pipeline ID to create the test result for. Only one of pipelineSlug or pipelineId is required.

    collectionMethod string

    Possible values: [manual, runner]

    The method used to collect the test runs

    branch string nullable

    The branch that the test result was created from

    commit string nullable

    The commit that the test result was created from

    name string nullable

    The name of the test result

    metadata object nullable
    property name* MetadataValueObject nullable
    oneOf
    type string required

    Possible values: [url]

    url uri required
    text string required
    type string required
    type string required

    Possible values: [url]

    url uri required
    text string required
    triggerRemoteEvals boolean

    Default value: true

    Optional flag to trigger remote evaluations

    testRuns object[] required
  • Array [
    • id uuid nullable

    The ID of the test run

    caseId uuid

    The ID of the test case. Mutually exclusive with 'name' and 'inputs'.

    name string

    Used for supplying local data. The name of the test run. Used with 'inputs' and mutually exclusive with 'caseId'.

    inputs object

    Used for supplying local data. The input data for the test run. Used with 'name' and mutually exclusive with 'caseId'.

    property name* any

    Used for supplying local data. The input data for the test run. Used with 'name' and mutually exclusive with 'caseId'.

    expectedOutputs object

    Used for supplying local data. The expected outputs for the test run. Used with 'name' and mutually exclusive with 'caseId'. Optional, since not all evaluators require expected outputs.

    property name* any

    Used for supplying local data. The expected outputs for the test run. Used with 'name' and mutually exclusive with 'caseId'. Optional, since not all evaluators require expected outputs.

    metadata object nullable
    property name* MetadataValueObject nullable
    oneOf
    type string required

    Possible values: [url]

    url uri required
    text string required
    type string required
    type string required

    Possible values: [url]

    url uri required
    text string required
    stepRuns object[] required

    Use outputs.steps insteads.

  • Array [
    • providerName string required
    invocation string required
    modelParams object required
    property name* any
    inputs object required
    property name* any
    outputs object required
    property name* any
    context object
    userId string
    render object
    type string required
    key string
    metadata object nullable
    property name* MetadataValueObject nullable
    oneOf
    type string required

    Possible values: [url]

    url uri required
    text string required
    type string required
    type string required

    Possible values: [url]

    url uri required
    text string required
    elapsedTime integer required
    startTime date-time required
    endTime date-time required
    error string nullable
  • ]
    • evaluations object[]

    Optional array of local evaluations

  • Array [
    • name string required

    The name of the local evaluation

    value number required

    The numeric value of the evaluation

    label string nullable

    Optional label for the evaluation

    debug object nullable
    resolvedPrompt string

    The resolved prompt used for the evaluation

    response string

    The response received from the evaluation

    finalClassification string

    The final classification of the evaluation

    processorLogs array[]

    Processor logs

    logs array[]

    Evaluator logs

    error object
    message string

    Error message

    date date-time

    Date and time of the error

  • ]
    • path object[] nullable

    Optional path which describes the organizational hierarchy of this test run in relation to all the pipeline tests.

  • Array [
    • type string required

    Possible values: [suite]

    name string required
  • ]
    • error string nullable
  • ]
Responses

Test result created successfully

Schema
    resultId uuid required
Loading...