Skip to content

On-premise users: click in-app to access the full platform documentation for your version of DataRobot.

Custom Model Vector Databases (GenAI)

This page outlines the operations, endpoints, parameters, and example requests and responses for the Custom Model Vector Databases (GenAI).

GET /api/v2/genai/customModelVectorDatabaseValidations/

List custom model vector database validations.

Code samples

# You can also use wget
curl -X GET https://app.datarobot.com/api/v2/genai/customModelVectorDatabaseValidations/ \
  -H "Accept: application/json" \
  -H "Authorization: Bearer {access-token}"

Parameters

Name In Type Required Description
useCaseId query any false Only retrieve the custom model vector database validations associated with these use case IDs.
playgroundId query string false Only retrieve the custom model vector database validations associated with this playground ID.
offset query integer false Skip the specified number of values.
limit query integer false Retrieve only the specified number of values.
search query any false Only retrieve the custom model vector database validations matching the search query.
sort query any false Apply this sort order to the results. Valid options are "name", "deploymentName", "userName", "creationDate". Prefix the attribute name with a dash to sort in descending order, e.g., sort=-creationDate.
completedOnly query boolean false If true, only retrieve the completed custom model vector database validations. The default is false.
deploymentId query string false Only retrieve the custom model vector database validations associated with this deployment ID.
modelId query string false Only retrieve the custom model vector database validations associated with this model ID.
promptColumnName query any false Only retrieve the custom model vector database validations where the custom model uses this column name for prompt input.
targetColumnName query any false Only retrieve the custom model vector database validations where the custom model uses this column name for prediction output.

Example responses

200 Response

{
  "count": 0,
  "data": [
    {
      "creationDate": "2019-08-24T14:15:22Z",
      "deploymentAccessData": {
        "authorizationHeader": "[REDACTED]",
        "datarobotKey": "string",
        "inputType": "CSV",
        "modelType": "TEXT_GENERATION",
        "predictionApiUrl": "string"
      },
      "deploymentId": "string",
      "deploymentName": "string",
      "errorMessage": "Unknown vector database error occurred",
      "id": "string",
      "modelId": "string",
      "name": "string",
      "predictionTimeout": 0,
      "promptColumnName": "string",
      "targetColumnName": "string",
      "tenantId": "string",
      "useCaseId": "string",
      "userId": "string",
      "userName": "string",
      "validationStatus": "TESTING"
    }
  ],
  "next": "string",
  "previous": "string",
  "totalCount": 0
}

Responses

Status Meaning Description Schema
200 OK Custom model vector database validations successfully retrieved. ListCustomModelVectorDatabaseValidationsResponse
422 Unprocessable Entity Validation Error HTTPValidationErrorResponse

To perform this operation, you must be authenticated by means of one of the following methods:

BearerAuth

POST /api/v2/genai/customModelVectorDatabaseValidations/

Validate a vector database hosted in a custom model deployment for use in the playground.

Code samples

# You can also use wget
curl -X POST https://app.datarobot.com/api/v2/genai/customModelVectorDatabaseValidations/ \
  -H "Content-Type: application/json" \
  -H "Accept: application/json" \
  -H "Authorization: Bearer {access-token}"

Body parameter

{
  "deploymentId": "string",
  "modelId": "string",
  "name": "Untitled",
  "predictionTimeout": 300,
  "promptColumnName": "string",
  "targetColumnName": "string",
  "useCaseId": "string"
}

Parameters

Name In Type Required Description
body body CreateCustomModelValidationRequest true none

Example responses

202 Response

{
  "creationDate": "2019-08-24T14:15:22Z",
  "deploymentAccessData": {
    "authorizationHeader": "[REDACTED]",
    "datarobotKey": "string",
    "inputType": "CSV",
    "modelType": "TEXT_GENERATION",
    "predictionApiUrl": "string"
  },
  "deploymentId": "string",
  "deploymentName": "string",
  "errorMessage": "Unknown vector database error occurred",
  "id": "string",
  "modelId": "string",
  "name": "string",
  "predictionTimeout": 0,
  "promptColumnName": "string",
  "targetColumnName": "string",
  "tenantId": "string",
  "useCaseId": "string",
  "userId": "string",
  "userName": "string",
  "validationStatus": "TESTING"
}

Responses

Status Meaning Description Schema
202 Accepted Custom model vector database validation job successfully accepted. Follow the Location header to poll for job execution status. CustomModelVectorDatabaseValidationResponse
422 Unprocessable Entity Validation Error HTTPValidationErrorResponse

To perform this operation, you must be authenticated by means of one of the following methods:

BearerAuth

DELETE /api/v2/genai/customModelVectorDatabaseValidations/{validationId}/

Delete an existing custom model vector database validation.

Code samples

# You can also use wget
curl -X DELETE https://app.datarobot.com/api/v2/genai/customModelVectorDatabaseValidations/{validationId}/ \
  -H "Accept: application/json" \
  -H "Authorization: Bearer {access-token}"

Parameters

Name In Type Required Description
validationId path string true The ID of the custom model vector database validation to delete.

Example responses

422 Response

{
  "detail": [
    {
      "loc": [
        "string"
      ],
      "msg": "string",
      "type": "string"
    }
  ]
}

Responses

Status Meaning Description Schema
204 No Content Successfully deleted custom model vector database validation None
422 Unprocessable Entity Validation Error HTTPValidationErrorResponse

To perform this operation, you must be authenticated by means of one of the following methods:

BearerAuth

GET /api/v2/genai/customModelVectorDatabaseValidations/{validationId}/

Retrieve the status of validating a custom model vector database.

Code samples

# You can also use wget
curl -X GET https://app.datarobot.com/api/v2/genai/customModelVectorDatabaseValidations/{validationId}/ \
  -H "Accept: application/json" \
  -H "Authorization: Bearer {access-token}"

Parameters

Name In Type Required Description
validationId path string true The ID of the custom model vector database validation to retrieve.

Example responses

200 Response

{
  "creationDate": "2019-08-24T14:15:22Z",
  "deploymentAccessData": {
    "authorizationHeader": "[REDACTED]",
    "datarobotKey": "string",
    "inputType": "CSV",
    "modelType": "TEXT_GENERATION",
    "predictionApiUrl": "string"
  },
  "deploymentId": "string",
  "deploymentName": "string",
  "errorMessage": "Unknown vector database error occurred",
  "id": "string",
  "modelId": "string",
  "name": "string",
  "predictionTimeout": 0,
  "promptColumnName": "string",
  "targetColumnName": "string",
  "tenantId": "string",
  "useCaseId": "string",
  "userId": "string",
  "userName": "string",
  "validationStatus": "TESTING"
}

Responses

Status Meaning Description Schema
200 OK Custom model vector database validation status successfully retrieved. CustomModelVectorDatabaseValidationResponse
422 Unprocessable Entity Validation Error HTTPValidationErrorResponse

To perform this operation, you must be authenticated by means of one of the following methods:

BearerAuth

PATCH /api/v2/genai/customModelVectorDatabaseValidations/{validationId}/

Edit an existing custom model vector database validation.

Code samples

# You can also use wget
curl -X PATCH https://app.datarobot.com/api/v2/genai/customModelVectorDatabaseValidations/{validationId}/ \
  -H "Content-Type: application/json" \
  -H "Accept: application/json" \
  -H "Authorization: Bearer {access-token}"

Body parameter

{
  "deploymentId": "string",
  "modelId": "string",
  "name": "string",
  "predictionTimeout": 1,
  "promptColumnName": "string",
  "targetColumnName": "string"
}

Parameters

Name In Type Required Description
validationId path string true The ID of the custom model vector database validation to edit.
body body EditCustomModelValidationRequest true none

Example responses

200 Response

{
  "creationDate": "2019-08-24T14:15:22Z",
  "deploymentAccessData": {
    "authorizationHeader": "[REDACTED]",
    "datarobotKey": "string",
    "inputType": "CSV",
    "modelType": "TEXT_GENERATION",
    "predictionApiUrl": "string"
  },
  "deploymentId": "string",
  "deploymentName": "string",
  "errorMessage": "Unknown vector database error occurred",
  "id": "string",
  "modelId": "string",
  "name": "string",
  "predictionTimeout": 0,
  "promptColumnName": "string",
  "targetColumnName": "string",
  "tenantId": "string",
  "useCaseId": "string",
  "userId": "string",
  "userName": "string",
  "validationStatus": "TESTING"
}

Responses

Status Meaning Description Schema
200 OK Custom model vector database validation successfully updated. CustomModelVectorDatabaseValidationResponse
422 Unprocessable Entity Validation Error HTTPValidationErrorResponse

To perform this operation, you must be authenticated by means of one of the following methods:

BearerAuth

POST /api/v2/genai/customModelVectorDatabaseValidations/{validationId}/revalidate/

Revalidate an existing custom model vector database.

Code samples

# You can also use wget
curl -X POST https://app.datarobot.com/api/v2/genai/customModelVectorDatabaseValidations/{validationId}/revalidate/ \
  -H "Accept: application/json" \
  -H "Authorization: Bearer {access-token}"

Parameters

Name In Type Required Description
validationId path string true The ID of the custom model vector database validation to revalidate.

Example responses

200 Response

{
  "creationDate": "2019-08-24T14:15:22Z",
  "deploymentAccessData": {
    "authorizationHeader": "[REDACTED]",
    "datarobotKey": "string",
    "inputType": "CSV",
    "modelType": "TEXT_GENERATION",
    "predictionApiUrl": "string"
  },
  "deploymentId": "string",
  "deploymentName": "string",
  "errorMessage": "Unknown vector database error occurred",
  "id": "string",
  "modelId": "string",
  "name": "string",
  "predictionTimeout": 0,
  "promptColumnName": "string",
  "targetColumnName": "string",
  "tenantId": "string",
  "useCaseId": "string",
  "userId": "string",
  "userName": "string",
  "validationStatus": "TESTING"
}

Responses

Status Meaning Description Schema
200 OK Custom model vector database successfully revalidated. CustomModelVectorDatabaseValidationResponse
422 Unprocessable Entity Validation Error HTTPValidationErrorResponse

To perform this operation, you must be authenticated by means of one of the following methods:

BearerAuth

Schemas

CreateCustomModelValidationRequest

{
  "deploymentId": "string",
  "modelId": "string",
  "name": "Untitled",
  "predictionTimeout": 300,
  "promptColumnName": "string",
  "targetColumnName": "string",
  "useCaseId": "string"
}

CreateCustomModelValidationRequest

Properties

Name Type Required Restrictions Description
deploymentId string true The ID of the custom model deployment.
modelId string¦null false The ID of the model used in the deployment.
name string false maxLength: 5000
The name to use for the validated custom model.
predictionTimeout integer false maximum: 1200
minimum: 1
The timeout in seconds for the prediction when validating a custom model. Defaults to 300.
promptColumnName string true maxLength: 5000
The name of the column the custom model uses for prompt text input.
targetColumnName string true maxLength: 5000
The name of the column the custom model uses for prediction output.
useCaseId string¦null false The ID of the use case to associate with the validated custom model.

CustomModelValidationStatus

"TESTING"

CustomModelValidationStatus

Properties

Name Type Required Restrictions Description
CustomModelValidationStatus string false Status of custom model validation.

Enumerated Values

Property Value
CustomModelValidationStatus [TESTING, PASSED, FAILED]

CustomModelVectorDatabaseValidationResponse

{
  "creationDate": "2019-08-24T14:15:22Z",
  "deploymentAccessData": {
    "authorizationHeader": "[REDACTED]",
    "datarobotKey": "string",
    "inputType": "CSV",
    "modelType": "TEXT_GENERATION",
    "predictionApiUrl": "string"
  },
  "deploymentId": "string",
  "deploymentName": "string",
  "errorMessage": "Unknown vector database error occurred",
  "id": "string",
  "modelId": "string",
  "name": "string",
  "predictionTimeout": 0,
  "promptColumnName": "string",
  "targetColumnName": "string",
  "tenantId": "string",
  "useCaseId": "string",
  "userId": "string",
  "userName": "string",
  "validationStatus": "TESTING"
}

CustomModelVectorDatabaseValidationResponse

Properties

Name Type Required Restrictions Description
creationDate string(date-time) true The creation date of the custom model validation (ISO 8601 formatted).
deploymentAccessData DeploymentAccessDataApiFormatted¦null true The parameters used for accessing the deployment.
deploymentId string true The ID of the custom model deployment.
deploymentName string¦null false The name of the custom model deployment.
errorMessage VectorDatabaseErrorMessages¦null true The error message associated with the validation error (if the validation failed).
id string true The ID of the custom model validation.
modelId string true The ID of the model used in the deployment.
name string true The name of the validated custom model.
predictionTimeout integer true The timeout in seconds for the prediction API used in this custom model validation.
promptColumnName string true The name of the column the custom model uses for prompt text input.
targetColumnName string true The name of the column the custom model uses for prediction output.
tenantId string(uuid4) true The ID of the tenant the custom model validation belongs to.
useCaseId string¦null true The ID of the use case associated with the validated custom model.
userId string true The ID of the user that created this custom model validation.
userName string¦null false The name of the user that created this custom model validation.
validationStatus CustomModelValidationStatus true The status of the custom model validation.

DeploymentAccessDataApiFormatted

{
  "authorizationHeader": "[REDACTED]",
  "datarobotKey": "string",
  "inputType": "CSV",
  "modelType": "TEXT_GENERATION",
  "predictionApiUrl": "string"
}

DeploymentAccessDataApiFormatted

Properties

Name Type Required Restrictions Description
authorizationHeader string false The Authorization header to use for the deployment.
datarobotKey string¦null true The server key associated with the prediction API.
inputType DeploymentInputType true The format of the input data.
modelType SupportedDeploymentType true The type of the target output the deployment produces.
predictionApiUrl string true The URL of the deployment's prediction API.

DeploymentInputType

"CSV"

DeploymentInputType

Properties

Name Type Required Restrictions Description
DeploymentInputType string false The format of the input data submitted to a DataRobot deployment.

Enumerated Values

Property Value
DeploymentInputType [CSV, JSON]

EditCustomModelValidationRequest

{
  "deploymentId": "string",
  "modelId": "string",
  "name": "string",
  "predictionTimeout": 1,
  "promptColumnName": "string",
  "targetColumnName": "string"
}

EditCustomModelValidationRequest

Properties

Name Type Required Restrictions Description
deploymentId string¦null false If specified, changes the ID of the deployment associated with this custom model validation.
modelId string¦null false If specified, changes the ID of the model associated with this custom model validation.
name string¦null false maxLength: 5000
minLength: 1
minLength: 1
If specified, renames the custom model validation to this value.
predictionTimeout integer¦null false maximum: 1200
minimum: 1
If specified, sets the timeout in seconds for the prediction when validating a custom model.
promptColumnName string¦null false maxLength: 5000
If specified, changes the name of the column that will be used to format the prompt text input for the custom model deployment.
targetColumnName string¦null false maxLength: 5000
If specified, changes the name of the column that will be used to extract the prediction response from the custom model deployment.

HTTPValidationErrorResponse

{
  "detail": [
    {
      "loc": [
        "string"
      ],
      "msg": "string",
      "type": "string"
    }
  ]
}

HTTPValidationErrorResponse

Properties

Name Type Required Restrictions Description
detail [ValidationError] false none

ListCustomModelValidationSortQueryParam

"name"

ListCustomModelValidationSortQueryParam

Properties

Name Type Required Restrictions Description
ListCustomModelValidationSortQueryParam string false Sort order values for listing custom model validations.

Enumerated Values

Property Value
ListCustomModelValidationSortQueryParam [name, -name, deploymentName, -deploymentName, userName, -userName, creationDate, -creationDate]

ListCustomModelVectorDatabaseValidationsResponse

{
  "count": 0,
  "data": [
    {
      "creationDate": "2019-08-24T14:15:22Z",
      "deploymentAccessData": {
        "authorizationHeader": "[REDACTED]",
        "datarobotKey": "string",
        "inputType": "CSV",
        "modelType": "TEXT_GENERATION",
        "predictionApiUrl": "string"
      },
      "deploymentId": "string",
      "deploymentName": "string",
      "errorMessage": "Unknown vector database error occurred",
      "id": "string",
      "modelId": "string",
      "name": "string",
      "predictionTimeout": 0,
      "promptColumnName": "string",
      "targetColumnName": "string",
      "tenantId": "string",
      "useCaseId": "string",
      "userId": "string",
      "userName": "string",
      "validationStatus": "TESTING"
    }
  ],
  "next": "string",
  "previous": "string",
  "totalCount": 0
}

ListCustomModelVectorDatabaseValidationsResponse

Properties

Name Type Required Restrictions Description
count integer true The number of records on this page.
data [CustomModelVectorDatabaseValidationResponse] true The list of records.
next string¦null true The URL to the next page, or null if there is no such page.
previous string¦null true The URL to the previous page, or null if there is no such page.
totalCount integer true The total number of records.

SupportedDeploymentType

"TEXT_GENERATION"

SupportedDeploymentType

Properties

Name Type Required Restrictions Description
SupportedDeploymentType string false The type of the target output a DataRobot deployment produces.

Enumerated Values

Property Value
SupportedDeploymentType [TEXT_GENERATION, UNSTRUCTURED, REGRESSION, MULTICLASS, BINARY, NOT_SUPPORTED]

ValidationError

{
  "loc": [
    "string"
  ],
  "msg": "string",
  "type": "string"
}

ValidationError

Properties

Name Type Required Restrictions Description
loc [anyOf] true none

anyOf

Name Type Required Restrictions Description
» anonymous string false none

or

Name Type Required Restrictions Description
» anonymous integer false none

continued

Name Type Required Restrictions Description
msg string true none
type string true none

VectorDatabaseErrorMessages

"Unknown vector database error occurred"

VectorDatabaseErrorMessages

Properties

Name Type Required Restrictions Description
VectorDatabaseErrorMessages string false Error messages for vector database errors.

Enumerated Values

Property Value
VectorDatabaseErrorMessages [Unknown vector database error occurred, A vector database prediction timeout error occurred., Downloading or finding embedding model weights failed, Loading dataset failed, All loaded documents are empty or contain no text, Raw loaded text dataset size exceeds the limit for the chosen embedding model, Splitting documents into text chunks failed, Generating embeddings from text chunks failed, Creating an index from embeddings failed, Storing vector database assets failed, Calcluating the size of stored vector database assets failed, API token was deleted or revoked, Document retrieval failed, Vector database was deleted, Loading the embedding model failed, External vector database connection is not available, External vector database deployment has been deleted, No access to external vector database deployment, Only unstructured deployment types are supported, An API token is required but was not provided in the request, Deployment prediction server did not accept the request, Deployment prediction does not comply with the expected format, API token was deleted or revoked, External embedding model connection is not available, External embedding deployment has been deleted, No access to external embedding deployment, Only unstructured deployment types are supported, An API token is required but was not provided in the request, Deployment prediction server did not accept the request, Deployment prediction does not comply with the expected format, Worker process was unexpectedly terminated]

Updated June 21, 2024