List tests

curl --request GET \
  --url https://api.avidoai.com/v0/tests \
  --header 'x-api-key: <api-key>' \
  --header 'x-application-id: <api-key>'

{
  "data": [
    {
      "id": "123e4567-e89b-12d3-a456-426614174000",
      "applicationId": "<string>",
      "createdAt": "2024-01-05T12:34:56.789Z",
      "modifiedAt": "2024-01-05T12:34:56.789Z",
      "status": "COMPLETED",
      "evals": [
        {
          "id": "123e4567-e89b-12d3-a456-426614174000",
          "orgId": "org_123456",
          "createdAt": "2024-01-05T12:34:56.789Z",
          "modifiedAt": "2024-01-05T12:34:56.789Z",
          "status": "COMPLETED",
          "results": {},
          "definition": {
            "id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
            "createdAt": "2024-01-05T12:34:56.789Z",
            "modifiedAt": "2024-01-05T12:34:56.789Z",
            "type": "NATURALNESS",
            "name": "<string>",
            "globalConfig": {
              "criterion": "<string>"
            },
            "styleGuideId": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
            "application": {
              "id": "123e4567-e89b-12d3-a456-426614174000",
              "orgId": "org_123456",
              "title": "Customer Support Bot",
              "slug": "customer-support-bot",
              "description": "AI assistant for customer support inquiries",
              "context": "You are a helpful customer support assistant...",
              "language": "en",
              "type": "CHATBOT",
              "environment": "DEV",
              "createdAt": "2024-01-05T12:34:56.789Z",
              "modifiedAt": "2024-01-05T12:34:56.789Z"
            }
          },
          "score": 0.9,
          "passed": true
        }
      ],
      "taskId": "123e4567-e89b-12d3-a456-426614174000",
      "runId": "123e4567-e89b-12d3-a456-426614174000",
      "task": {
        "id": "123e4567-e89b-12d3-a456-426614174000",
        "title": "Credit Card Block",
        "topicId": "789e4567-e89b-12d3-a456-426614174000"
      },
      "traceId": "789e4567-e89b-12d3-a456-426614174000",
      "result": {
        "passed": 1,
        "failed": 1,
        "total": 1,
        "passRate": 50,
        "averageScore": 0.5
      }
    }
  ],
  "pagination": {
    "skip": 0,
    "limit": 25,
    "total": 100,
    "totalPages": 4
  }
}

GET

tests

List tests

curl --request GET \
  --url https://api.avidoai.com/v0/tests \
  --header 'x-api-key: <api-key>' \
  --header 'x-application-id: <api-key>'

{
  "data": [
    {
      "id": "123e4567-e89b-12d3-a456-426614174000",
      "applicationId": "<string>",
      "createdAt": "2024-01-05T12:34:56.789Z",
      "modifiedAt": "2024-01-05T12:34:56.789Z",
      "status": "COMPLETED",
      "evals": [
        {
          "id": "123e4567-e89b-12d3-a456-426614174000",
          "orgId": "org_123456",
          "createdAt": "2024-01-05T12:34:56.789Z",
          "modifiedAt": "2024-01-05T12:34:56.789Z",
          "status": "COMPLETED",
          "results": {},
          "definition": {
            "id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
            "createdAt": "2024-01-05T12:34:56.789Z",
            "modifiedAt": "2024-01-05T12:34:56.789Z",
            "type": "NATURALNESS",
            "name": "<string>",
            "globalConfig": {
              "criterion": "<string>"
            },
            "styleGuideId": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
            "application": {
              "id": "123e4567-e89b-12d3-a456-426614174000",
              "orgId": "org_123456",
              "title": "Customer Support Bot",
              "slug": "customer-support-bot",
              "description": "AI assistant for customer support inquiries",
              "context": "You are a helpful customer support assistant...",
              "language": "en",
              "type": "CHATBOT",
              "environment": "DEV",
              "createdAt": "2024-01-05T12:34:56.789Z",
              "modifiedAt": "2024-01-05T12:34:56.789Z"
            }
          },
          "score": 0.9,
          "passed": true
        }
      ],
      "taskId": "123e4567-e89b-12d3-a456-426614174000",
      "runId": "123e4567-e89b-12d3-a456-426614174000",
      "task": {
        "id": "123e4567-e89b-12d3-a456-426614174000",
        "title": "Credit Card Block",
        "topicId": "789e4567-e89b-12d3-a456-426614174000"
      },
      "traceId": "789e4567-e89b-12d3-a456-426614174000",
      "result": {
        "passed": 1,
        "failed": 1,
        "total": 1,
        "passRate": 50,
        "averageScore": 0.5
      }
    }
  ],
  "pagination": {
    "skip": 0,
    "limit": 25,
    "total": 100,
    "totalPages": 4
  }
}

Authorizations

x-api-key

string

header

required

Your unique Avido API key

x-application-id

string

header

required

Your unique Avido Application ID

Query Parameters

skip

integer

default:0

Number of items to skip before starting to collect the result set.

Required range: x >= 0

Example:

0

limit

integer

default:25

Number of items to include in the result set.

Required range: 1 <= x <= 100

Example:

25

orderBy

string

default:createdAt

Field to order by in the result set.

Example:

"createdAt"

orderDir

enum<string>

default:desc

Order direction.

Available options:

asc,

desc

Example:

"desc"

status

enum<string>[]

Filter by test status (e.g. COMPLETED, FAILED)

Status of the evaluation/test

Available options:

PENDING,

IN_PROGRESS,

COMPLETED,

FAILED

Example:

["COMPLETED", "FAILED"]

startDate

string<date-time>

Filter eval runs after this date (inclusive).

Example:

"2024-01-01T00:00:00.000Z"

endDate

string<date-time>

Filter eval runs before this date (inclusive).

Example:

"2024-12-31T23:59:59.999Z"

passRateStatuses

enum<string>[]

Filter by pass rate status badges (success: >75%, warning: 51-75%, error: ≤50%)

Available options:

success,

warning,

error

Example:

["success", "warning"]

evalDefinitionId

string<uuid>[]

Filter tests by eval definition ID

Example:

"123e4567-e89b-12d3-a456-426614174000"

experimentVariantId

string<uuid>[]

Filter tests by experiment variant ID

Example:

"123e4567-e89b-12d3-a456-426614174000"

runType

enum<string>[]

Filter tests by run type (MANUAL, SCHEDULED, EXPERIMENT)

Available options:

MANUAL,

SCHEDULED,

EXPERIMENT

Example:

["MANUAL", "SCHEDULED"]

Response

Successfully retrieved evaluations

Paginated response containing tests

data

Test · object[]

required

Show child attributes

data.id

string<uuid>

required

Unique identifier of the run

Example:

"123e4567-e89b-12d3-a456-426614174000"

data.applicationId

string

required

The ID of the application this test belongs to

data.createdAt

string<date-time>

required

When the test was created

Example:

"2024-01-05T12:34:56.789Z"

data.modifiedAt

string<date-time>

required

When the test was last modified

Example:

"2024-01-05T12:34:56.789Z"

data.status

enum<string>

required

Status of the evaluation/test

Available options:

PENDING,

IN_PROGRESS,

COMPLETED,

FAILED

Example:

"COMPLETED"

data.evals

Eval · object[]

required

Array of evaluations in this run

Show child attributes

data.evals.id

string<uuid>

required

Unique identifier of the evaluation

Example:

"123e4567-e89b-12d3-a456-426614174000"

data.evals.orgId

string

required

Organization ID that owns this evaluation

Example:

"org_123456"

data.evals.createdAt

string<date-time>

required

When the evaluation was created

Example:

"2024-01-05T12:34:56.789Z"

data.evals.modifiedAt

string<date-time>

required

When the evaluation was last modified

Example:

"2024-01-05T12:34:56.789Z"

data.evals.status

enum<string>

required

Status of the evaluation/test

Available options:

PENDING,

IN_PROGRESS,

COMPLETED,

FAILED

Example:

"COMPLETED"

data.evals.results

object

required

Results of the evaluation (structure depends on eval type).

Show child attributes

data.evals.results.{key}

any

data.evals.definition

object

required

Show child attributes

data.evals.definition.id

string<uuid>

required

data.evals.definition.createdAt

string<date-time>

required

When the eval definition was created

Example:

"2024-01-05T12:34:56.789Z"

data.evals.definition.modifiedAt

string<date-time>

required

When the eval definition was last modified

Example:

"2024-01-05T12:34:56.789Z"

data.evals.definition.type

enum<string>

required

The type of evaluation

Available options:

NATURALNESS,

STYLE,

RECALL,

CUSTOM,

FACT,

OUTPUT_MATCH

Example:

"NATURALNESS"

data.evals.definition.name

string

required

data.evals.definition.globalConfig

object

Option 1
Option 2
Option 3
Option 4

Show child attributes

data.evals.definition.globalConfig.criterion

string

required

The criterion describes what our evaluation LLM must look for in the response. Remember that the answer to the criterion must be as a pass/fail.

Minimum string length: 1

data.evals.definition.styleGuideId

string<uuid> | null

data.evals.definition.application

Application · object

Application configuration and metadata

Show child attributes

data.evals.definition.application.id

string<uuid>

required

Unique identifier of the application

Example:

"123e4567-e89b-12d3-a456-426614174000"

data.evals.definition.application.orgId

string

required

Organization ID that owns this application

Example:

"org_123456"

data.evals.definition.application.title

string

required

Title of the application

Minimum string length: 1

Example:

"Customer Support Bot"

data.evals.definition.application.slug

string

required

URL-friendly slug for the application

Minimum string length: 1

Example:

"customer-support-bot"

data.evals.definition.application.description

string

required

Description of the application

Example:

"AI assistant for customer support inquiries"

data.evals.definition.application.context

string

required

Context/instructions for the application

Example:

"You are a helpful customer support assistant..."

data.evals.definition.application.language

string

required

Language of the application

Example:

"en"

data.evals.definition.application.type

enum<string>

required

Type of the application. Valid values are CHATBOT or AGENT.

Available options:

CHATBOT,

AGENT

Example:

"CHATBOT"

data.evals.definition.application.environment

enum<string>

default:DEV

required

Environment of the application. Defaults to DEV.

Available options:

DEV,

PROD

Example:

"DEV"

data.evals.definition.application.createdAt

string<date-time>

required

When the application was created

Example:

"2024-01-05T12:34:56.789Z"

data.evals.definition.application.modifiedAt

string<date-time>

required

When the application was last modified

Example:

"2024-01-05T12:34:56.789Z"

data.evals.score

number

required

Overall score of the evaluation

Example:

0.9

data.evals.passed

boolean

required

Whether the evaluation passed

Example:

true

data.taskId

string<uuid>

required

The unique identifier of the task

Example:

"123e4567-e89b-12d3-a456-426614174000"

data.runId

string<uuid>

required

The unique identifier of the run

Example:

"123e4567-e89b-12d3-a456-426614174000"

data.task

Task · object

required

A task that represents a specific job-to-be-done by the LLM in the user application.

Show child attributes

data.task.id

string<uuid>

required

The unique identifier of the task

Example:

"123e4567-e89b-12d3-a456-426614174000"

data.task.title

string

required

The title of the task

Example:

"Credit Card Block"

data.task.topicId

string<uuid>

The ID of the topic this task belongs to

Example:

"789e4567-e89b-12d3-a456-426614174000"

data.traceId

string

Optional ID of the trace this run is associated with

Example:

"789e4567-e89b-12d3-a456-426614174000"

data.result

object

Aggregated test result with pass/fail statistics

Show child attributes

data.result.passed

integer

required

Required range: x >= 0

data.result.failed

integer

required

Required range: x >= 0

data.result.total

integer

required

Required range: x >= 0

data.result.passRate

number

required

Required range: 0 <= x <= 100

data.result.averageScore

number

required

Required range: 0 <= x <= 1

pagination

PaginationResponse · object

required

Pagination metadata returned in a paginated response.

Show child attributes

pagination.skip

integer

required

Number of items skipped.

Required range: x >= 0

Example:

0

pagination.limit

integer

required

Number of items per page.

Required range: x >= 1

Example:

25

pagination.total

integer

required

Total number of items available.

Required range: x >= 0

Example:

100

pagination.totalPages

integer

required

Total number of pages available.

Required range: x >= 0

Example:

4

Create an inference step Get a single test by ID

⌘I

Webhook

Applications

Threads

Ingestion

Tasks

Task Schedules

Eval Definitions

Experiments

Inference Steps

Evals

Tests

Topics

Annotations

Issues

Runs

Style Guides

Documents

Document Chunks

Document Versions

Tags

Document Tags

Task Tags

Scrape Jobs

Quickstarts

List tests

Authorizations

Query Parameters

Response