Batch Create

POSThttps://api.confident-ai.com/v1/metrics/batch

Creates a batch of new metrics on your Confident AI project.

POST/v1/metrics/batch

curl -X POST "https://api.confident-ai.com/v1/metrics/batch" \
  -H "CONFIDENT_API_KEY: <PROJECT-API-KEY>" \
  -H "Content-Type: application/json" \
  -d '{
  "metrics": [
    {
      "name": "Correctness",
      "criteria": "Determine if the `actual output` is correct based on the `expected output`.",
      "evaluationParams": [
        "actualOutput",
        "expectedOutput"
      ],
      "multiTurn": false
    }
  ]
}'

200

{
  "success": true,
  "data": {
    "ids": [
      "METRIC-ID"
    ]
  },
  "deprecated": false
}

Headers

CONFIDENT_API_KEYstringRequired
The API key of your Confident AI project.

Request body

metricslist of objectsRequired
This is the list of metrics you want to add to your project.
Show 7 propertiesHide 7 properties
- idstringRequired
  This is the unique id of the metric.
- namestringRequired
  This is the name of the metric, it's unique to all metrics.
- criteriastring
  This is the criteria this metric uses to evaluate test cases.
- evaluationStepslist of strings
  An alternative to criteria — a list of steps used to evaluate test cases.
- requiredParameterslist of enums | list of enumsRequired
  The parameters required by this metric for evaluation.
  Show 2 variantsHide 2 variants
  - singleTurnRequiredParameterslist of enums
    Show 9 enum valuesHide 9 enum values
    input
    actualOutput
    expectedOutput
    context
    expectedTools
    toolsCalled
    retrievalContext
    metadata
    tags
  - OR
  - multiTurnRequiredParameterslist of enums
    Show 8 enum valuesHide 8 enum values
    content
    role
    scenario
    expectedOutcome
    toolsCalled
    retrievalContext
    metadata
    tags
- multiTurnbooleanRequired
  This is true if the metric is used to evaluate multi-turn test cases.
- rubriclist of objects
  A list of score ranges (0–10 inclusive). Must be in order and non-overlapping. Click here to learn more
  Show 2 propertiesHide 2 properties
  - scoreRangelist of numbersRequired
    An array consisting of the ranges of scores to generate.
  - expectedOutcomestringRequired
    The expected outcome for your evaluation.

Response

successboolean
This is true if the metrics were created successfully.
dataobject
This maps to the ids of the metrics created.
Show 1 propertyHide 1 property
- idslist of strings
  The ids of the batch metrics created.
deprecatedboolean
This is true if this batch metrics endpoint is deprecated.

Built byConfident AI