Skip to content

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

DataRobot API resources > API reference documentation > DataRobot REST API > Chat Prompts (GenAI)

Chat Prompts (GenAI)

This page outlines the operations, endpoints, parameters, and example requests and responses for the Chat Prompts (GenAI).

GET /api/v2/genai/chatPrompts/

List chat prompts.

Code samples

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

Parameters

Name In Type Required Description
playgroundId query string false Only retrieve the chat prompts associated with this playground ID.
llmBlueprintId query string false Only retrieve the chat prompts associated with this LLM blueprint ID. If specified, will retrieve the chat prompts for the oldest chat in this LLM blueprint.
chatId query string false Only retrieve the chat prompts associated with this chat ID.
offset query integer false Skip the specified number of values.
limit query integer false Retrieve only the specified number of values.

Example responses

200 Response

{
  "count": 0,
  "data": [
    {
      "chatContextId": "string",
      "chatId": "string",
      "chatPromptIdsIncludedInHistory": [
        "string"
      ],
      "citations": [
        {
          "page": 0,
          "source": "string",
          "text": "string"
        }
      ],
      "confidenceScores": {
        "bleu": 0,
        "meteor": 0,
        "rouge": 0
      },
      "creationDate": "2019-08-24T14:15:22Z",
      "creationUserId": "string",
      "executionStatus": "NEW",
      "id": "string",
      "llmBlueprintId": "string",
      "llmId": "azure-openai-gpt-3.5-turbo",
      "llmSettings": {
        "maxCompletionLength": 0,
        "systemPrompt": "string",
        "temperature": 0,
        "topP": 0
      },
      "resultMetadata": {
        "cost": 0,
        "errorMessage": "string",
        "estimatedDocsTokenCount": 0,
        "feedbackResult": {
          "negativeUserIds": [],
          "positiveUserIds": []
        },
        "finalPrompt": "string",
        "inputTokenCount": 0,
        "latencyMilliseconds": 0,
        "metrics": [],
        "outputTokenCount": 0,
        "totalTokenCount": 0
      },
      "resultText": "string",
      "text": "string",
      "userName": "string",
      "vectorDatabaseId": "string",
      "vectorDatabaseSettings": {
        "maxDocumentsRetrievedPerPrompt": 10,
        "maxTokens": 0
      }
    }
  ],
  "next": "string",
  "previous": "string",
  "totalCount": 0
}

Responses

Status Meaning Description Schema
200 OK Successful Response ListChatPromptsResponse
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/chatPrompts/

Request the execution of a new prompt within a chat or an LLM blueprint.

Code samples

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

Body parameter

{
  "chatId": "string",
  "llmBlueprintId": "string",
  "llmId": "azure-openai-gpt-3.5-turbo",
  "llmSettings": {
    "maxCompletionLength": 0,
    "systemPrompt": "string",
    "temperature": 0,
    "topP": 0
  },
  "text": "string",
  "vectorDatabaseId": "string",
  "vectorDatabaseSettings": {
    "maxDocumentsRetrievedPerPrompt": 10,
    "maxTokens": 0
  }
}

Parameters

Name In Type Required Description
body body CreateChatPromptRequest true none

Example responses

202 Response

{
  "chatContextId": "string",
  "chatId": "string",
  "chatPromptIdsIncludedInHistory": [
    "string"
  ],
  "citations": [
    {
      "page": 0,
      "source": "string",
      "text": "string"
    }
  ],
  "confidenceScores": {
    "bleu": 0,
    "meteor": 0,
    "rouge": 0
  },
  "creationDate": "2019-08-24T14:15:22Z",
  "creationUserId": "string",
  "executionStatus": "NEW",
  "id": "string",
  "llmBlueprintId": "string",
  "llmId": "azure-openai-gpt-3.5-turbo",
  "llmSettings": {
    "maxCompletionLength": 0,
    "systemPrompt": "string",
    "temperature": 0,
    "topP": 0
  },
  "resultMetadata": {
    "cost": 0,
    "errorMessage": "string",
    "estimatedDocsTokenCount": 0,
    "feedbackResult": {
      "negativeUserIds": [],
      "positiveUserIds": []
    },
    "finalPrompt": "string",
    "inputTokenCount": 0,
    "latencyMilliseconds": 0,
    "metrics": [],
    "outputTokenCount": 0,
    "totalTokenCount": 0
  },
  "resultText": "string",
  "text": "string",
  "userName": "string",
  "vectorDatabaseId": "string",
  "vectorDatabaseSettings": {
    "maxDocumentsRetrievedPerPrompt": 10,
    "maxTokens": 0
  }
}

Responses

Status Meaning Description Schema
202 Accepted Successful Response ChatPromptResponse
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/chatPrompts/{chatPromptId}/

Delete an existing chat prompt.

Code samples

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

Parameters

Name In Type Required Description
chatPromptId path string true The ID of the chat prompt to delete.

Example responses

422 Response

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

Responses

Status Meaning Description Schema
204 No Content Successful Response 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/chatPrompts/{chatPromptId}/

Retrieve an existing chat prompt.

Code samples

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

Parameters

Name In Type Required Description
chatPromptId path string true The ID of the chat prompt to retrieve.

Example responses

200 Response

{
  "chatContextId": "string",
  "chatId": "string",
  "chatPromptIdsIncludedInHistory": [
    "string"
  ],
  "citations": [
    {
      "page": 0,
      "source": "string",
      "text": "string"
    }
  ],
  "confidenceScores": {
    "bleu": 0,
    "meteor": 0,
    "rouge": 0
  },
  "creationDate": "2019-08-24T14:15:22Z",
  "creationUserId": "string",
  "executionStatus": "NEW",
  "id": "string",
  "llmBlueprintId": "string",
  "llmId": "azure-openai-gpt-3.5-turbo",
  "llmSettings": {
    "maxCompletionLength": 0,
    "systemPrompt": "string",
    "temperature": 0,
    "topP": 0
  },
  "resultMetadata": {
    "cost": 0,
    "errorMessage": "string",
    "estimatedDocsTokenCount": 0,
    "feedbackResult": {
      "negativeUserIds": [],
      "positiveUserIds": []
    },
    "finalPrompt": "string",
    "inputTokenCount": 0,
    "latencyMilliseconds": 0,
    "metrics": [],
    "outputTokenCount": 0,
    "totalTokenCount": 0
  },
  "resultText": "string",
  "text": "string",
  "userName": "string",
  "vectorDatabaseId": "string",
  "vectorDatabaseSettings": {
    "maxDocumentsRetrievedPerPrompt": 10,
    "maxTokens": 0
  }
}

Responses

Status Meaning Description Schema
200 OK Successful Response ChatPromptResponse
422 Unprocessable Entity Validation Error HTTPValidationErrorResponse

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

BearerAuth

Schemas

ChatPromptResponse

{
  "chatContextId": "string",
  "chatId": "string",
  "chatPromptIdsIncludedInHistory": [
    "string"
  ],
  "citations": [
    {
      "page": 0,
      "source": "string",
      "text": "string"
    }
  ],
  "confidenceScores": {
    "bleu": 0,
    "meteor": 0,
    "rouge": 0
  },
  "creationDate": "2019-08-24T14:15:22Z",
  "creationUserId": "string",
  "executionStatus": "NEW",
  "id": "string",
  "llmBlueprintId": "string",
  "llmId": "azure-openai-gpt-3.5-turbo",
  "llmSettings": {
    "maxCompletionLength": 0,
    "systemPrompt": "string",
    "temperature": 0,
    "topP": 0
  },
  "resultMetadata": {
    "cost": 0,
    "errorMessage": "string",
    "estimatedDocsTokenCount": 0,
    "feedbackResult": {
      "negativeUserIds": [],
      "positiveUserIds": []
    },
    "finalPrompt": "string",
    "inputTokenCount": 0,
    "latencyMilliseconds": 0,
    "metrics": [],
    "outputTokenCount": 0,
    "totalTokenCount": 0
  },
  "resultText": "string",
  "text": "string",
  "userName": "string",
  "vectorDatabaseId": "string",
  "vectorDatabaseSettings": {
    "maxDocumentsRetrievedPerPrompt": 10,
    "maxTokens": 0
  }
}

ChatPromptResponse

Properties

Name Type Required Restrictions Description
chatContextId string¦null false The ID of the chat context for this prompt.
chatId string¦null false The ID of the chat this chat prompt belongs to.
chatPromptIdsIncludedInHistory [string]¦null false The list of IDs of the chat prompts included in this prompt's history.
citations [Citation] true The list of relevant vector database citations (in case of using a vector database).
confidenceScores ConfidenceScores¦null true The confidence scores that measure the similarity between the prompt context and the prompt completion.
creationDate string(date-time) true The creation date of the chat prompt (ISO 8601 formatted).
creationUserId string true The ID of the user that created the chat prompt.
executionStatus ExecutionStatus true The execution status of the chat prompt.
id string true The ID of the chat prompt.
llmBlueprintId string true The ID of the LLM blueprint the chat prompt belongs to.
llmId LanguageModelTypeId true The ID of the LLM used by the chat prompt.
llmSettings any false A key/value dictionary of LLM settings.

anyOf

Name Type Required Restrictions Description
» anonymous CommonLLMSettings false The settings that are available for all non-custom LLMs.

or

Name Type Required Restrictions Description
» anonymous CustomModelLLMSettings false The settings that are available for custom model LLMs.

continued

Name Type Required Restrictions Description
resultMetadata ResultMetadata¦null true The additional information about the chat prompt results.
resultText string¦null true The text of the prompt completion.
text string true The text of the user prompt.
userName string true The name of the user that created the chat prompt.
vectorDatabaseId string¦null false The ID of the vector database linked to this LLM blueprint.
vectorDatabaseSettings VectorDatabaseSettings¦null false A key/value dictionary of vector database settings.

Citation

{
  "page": 0,
  "source": "string",
  "text": "string"
}

Citation

Properties

Name Type Required Restrictions Description
page integer¦null false The source page number where the citation was found.
source string¦null true The source of the citation (e.g., a filename in the original dataset).
text string true The text of the citation.

CommonLLMSettings

{
  "maxCompletionLength": 0,
  "systemPrompt": "string",
  "temperature": 0,
  "topP": 0
}

CommonLLMSettings

Properties

Name Type Required Restrictions Description
maxCompletionLength integer¦null false Maximum number of tokens allowed in the completion. The combined count of this value and prompt tokens must be below the model's maximum context size, where prompt token count is comprised of system prompt, user prompt, recent chat history, and vector database citations.
systemPrompt string¦null false maxLength: 500000
 System prompt guides the style of the LLM response. It is a "universal" prompt, prepended to all individual prompts.
temperature number¦null false Temperature controls the randomness of model output, where higher values return more diverse output and lower values return more deterministic results.
topP number¦null false Top P sets a threshold that controls the selection of words included in the response, based on a cumulative probability cutoff for token selection. For example, 0.2 considers only the top 20% probability mass. Higher numbers return more diverse options for outputs.

ConfidenceScores

{
  "bleu": 0,
  "meteor": 0,
  "rouge": 0
}

ConfidenceScores

Properties

Name Type Required Restrictions Description
bleu number true BLEU score.
meteor number true METEOR score.
rouge number true ROUGE score.

CreateChatPromptRequest

{
  "chatId": "string",
  "llmBlueprintId": "string",
  "llmId": "azure-openai-gpt-3.5-turbo",
  "llmSettings": {
    "maxCompletionLength": 0,
    "systemPrompt": "string",
    "temperature": 0,
    "topP": 0
  },
  "text": "string",
  "vectorDatabaseId": "string",
  "vectorDatabaseSettings": {
    "maxDocumentsRetrievedPerPrompt": 10,
    "maxTokens": 0
  }
}

CreateChatPromptRequest

Properties

Name Type Required Restrictions Description
chatId string¦null false The ID of the chat this prompt belongs to. If LLM and vector database settings are not specified in the request, then the prompt will use the current settings of the chat.
llmBlueprintId string¦null false The ID of the LLM blueprint this prompt belongs to. If LLM and vector database settings are not specified in the request, then the prompt will use the current settings of the LLM blueprint.
llmId LanguageModelTypeId¦null false If specified, uses this LLM ID for the prompt and updates the settings of the corresponding chat or LLM blueprint to use this LLM ID.
llmSettings any false If specified, uses these LLM settings for the prompt and updates the settings of the corresponding chat or LLM blueprint to use these LLM settings.

anyOf

Name Type Required Restrictions Description
» anonymous CommonLLMSettings false The settings that are available for all non-custom LLMs.

or

Name Type Required Restrictions Description
» anonymous CustomModelLLMSettings false The settings that are available for custom model LLMs.

continued

Name Type Required Restrictions Description
text string true maxLength: 500000
 The text of the user prompt.
vectorDatabaseId string¦null false If specified, uses this vector database ID for the prompt and updates the settings of the corresponding chat or LLM blueprint to use this vector database ID.
vectorDatabaseSettings VectorDatabaseSettings¦null false If specified, uses these vector database settings for the prompt and updates the settings of the corresponding chat or LLM blueprint to use these vector database settings.

CustomModelLLMSettings

{
  "systemPrompt": "string",
  "validationId": "string"
}

CustomModelLLMSettings

Properties

Name Type Required Restrictions Description
systemPrompt string¦null false maxLength: 500000
 System prompt guides the style of the LLM response. It is a "universal" prompt, prepended to all individual prompts.
validationId string¦null false The validation ID of the custom model LLM.

ExecutionStatus

"NEW"

ExecutionStatus

Properties

Name Type Required Restrictions Description
ExecutionStatus string false Job execution status.

Enumerated Values

Property Value
ExecutionStatus [NEW, RUNNING, COMPLETED, ERROR]

FeedbackResult

{
  "negativeUserIds": [],
  "positiveUserIds": []
}

FeedbackResult

Properties

Name Type Required Restrictions Description
negativeUserIds [string] false The list of user IDs whose feedback is negative.
positiveUserIds [string] false The list of user IDs whose feedback is positive.

HTTPValidationErrorResponse

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

HTTPValidationErrorResponse

Properties

Name Type Required Restrictions Description
detail [ValidationError] false none

LanguageModelTypeId

"azure-openai-gpt-3.5-turbo"

LanguageModelTypeId

Properties

Name Type Required Restrictions Description
LanguageModelTypeId string false The ID that defines the type of the LLM.

Enumerated Values

Property Value
LanguageModelTypeId [azure-openai-gpt-3.5-turbo, azure-openai-gpt-3.5-turbo-16k, azure-openai-gpt-4, azure-openai-gpt-4-32k, amazon-titan, anthropic-claude-2, google-bison, custom-model]

ListChatPromptsResponse

{
  "count": 0,
  "data": [
    {
      "chatContextId": "string",
      "chatId": "string",
      "chatPromptIdsIncludedInHistory": [
        "string"
      ],
      "citations": [
        {
          "page": 0,
          "source": "string",
          "text": "string"
        }
      ],
      "confidenceScores": {
        "bleu": 0,
        "meteor": 0,
        "rouge": 0
      },
      "creationDate": "2019-08-24T14:15:22Z",
      "creationUserId": "string",
      "executionStatus": "NEW",
      "id": "string",
      "llmBlueprintId": "string",
      "llmId": "azure-openai-gpt-3.5-turbo",
      "llmSettings": {
        "maxCompletionLength": 0,
        "systemPrompt": "string",
        "temperature": 0,
        "topP": 0
      },
      "resultMetadata": {
        "cost": 0,
        "errorMessage": "string",
        "estimatedDocsTokenCount": 0,
        "feedbackResult": {
          "negativeUserIds": [],
          "positiveUserIds": []
        },
        "finalPrompt": "string",
        "inputTokenCount": 0,
        "latencyMilliseconds": 0,
        "metrics": [],
        "outputTokenCount": 0,
        "totalTokenCount": 0
      },
      "resultText": "string",
      "text": "string",
      "userName": "string",
      "vectorDatabaseId": "string",
      "vectorDatabaseSettings": {
        "maxDocumentsRetrievedPerPrompt": 10,
        "maxTokens": 0
      }
    }
  ],
  "next": "string",
  "previous": "string",
  "totalCount": 0
}

ListChatPromptsResponse

Properties

Name Type Required Restrictions Description
count integer true The number of records on this page.
data [ChatPromptResponse] 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.

MetricMetadata

{
  "costConfigurationId": "string",
  "customModelId": "string",
  "evaluationDatasetConfigurationId": "string",
  "formattedName": "string",
  "formattedValue": "string",
  "name": "string",
  "sidecarModelMetricValidationId": "string",
  "value": null
}

MetricMetadata

Properties

Name Type Required Restrictions Description
costConfigurationId string¦null false The ID of the cost configuration.
customModelId string¦null false The ID of the custom model used for the metric.
evaluationDatasetConfigurationId string¦null false The ID of the evaluation dataset configuration.
formattedName string¦null false The formatted name of the metric.
formattedValue string¦null false The formatted value of the metric.
name string true The name of the metric.
sidecarModelMetricValidationId string¦null false The validation ID of the sidecar model validation(in case of using a sidecar model deployment for the metric).
value any true The value of the metric.

ResultMetadata

{
  "cost": 0,
  "errorMessage": "string",
  "estimatedDocsTokenCount": 0,
  "feedbackResult": {
    "negativeUserIds": [],
    "positiveUserIds": []
  },
  "finalPrompt": "string",
  "inputTokenCount": 0,
  "latencyMilliseconds": 0,
  "metrics": [],
  "outputTokenCount": 0,
  "totalTokenCount": 0
}

ResultMetadata

Properties

Name Type Required Restrictions Description
cost number¦null false The estimated cost of executing the prompt.
errorMessage string¦null false The error message for the prompt (in case of an errored prompt).
estimatedDocsTokenCount integer false The estimated number of tokens in the documents retrieved from the vector database.
feedbackResult FeedbackResult false The user feedback associated with the prompt.
finalPrompt any false The final representation of the prompt that was submitted to the LLM.

anyOf

Name Type Required Restrictions Description
» anonymous string false none

or

Name Type Required Restrictions Description
» anonymous [object] false none
»» additionalProperties string false none

or

Name Type Required Restrictions Description
» anonymous object false none
»» additionalProperties any false none

anyOf

Name Type Required Restrictions Description
»»» anonymous string false none

or

Name Type Required Restrictions Description
»»» anonymous [object] false none
»»»» additionalProperties string false none

continued

Name Type Required Restrictions Description
inputTokenCount integer false The number of tokens in the LLM input. This number includes the tokens in the system prompt, the user prompt, the chat history (for history-aware chats) and the documents retrieved from the vector database (in case of using a vector database).
latencyMilliseconds integer true The latency of the LLM response (in milliseconds).
metrics [MetricMetadata] false The evaluation metrics for the prompt.
outputTokenCount integer false The number of tokens in the LLM output.
totalTokenCount integer false The combined number of tokens in the LLM input and output.

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

VectorDatabaseSettings

{
  "maxDocumentsRetrievedPerPrompt": 10,
  "maxTokens": 0
}

VectorDatabaseSettings

Properties

Name Type Required Restrictions Description
maxDocumentsRetrievedPerPrompt integer¦null false maximum: 10
 The maximum number of documents to retrieve from the vector database.
maxTokens integer¦null false The maximum number of tokens to retrieve from the vector database.
Updated September 3, 2024