Create conversation title

Available in VPC

LLM generates titles for conversations with users.

Request

This section describes the request format. The method and URI are as follows:

Request headers

For information about the headers common to all RAG APIs, see RAG request headers.

Request path parameters

The following describes the parameters.

Request body

You can include the following data in the body of your request:

messages

This is an array of messages that provide context for the conversation.

Request example

The request example is as follows:

curl --location --request POST 'https://kr-pub-gateway.rag.naverncp.com/api/v1/svc/{serviceId}/conversation-title' \
--header 'Authorization: Bearer {apiKey}' \
--header 'Accept: text/event-stream' \
--data '{
  "serviceId": "{serviceId}",
  "messages": [
    {
      "role": "user",
      "content": "How can I use AI safely?"
    },
    {
      "role": "assistant",
      "content": "The following describes how to use AI safely:\n\n1. **Technical improvements**: Provide accurate and neutral data to ensure that AI systems are not trained with misinformation or bias, and continue research and development to improve the stability and reliability of the system.\n\n2. **Ethical guidelines and adherence**: Like NAVER, companies should create and adhere to their own AI ethics guidelines. For example, emphasize human-centered values and have an AI ethics consultation process so that all employees can consider and discuss ethical issues.\n\n3. **Education and training**: Educate those who use AI technology about its proper use and potential risks to help prevent over-reliance on AI and maintain decision-making control.\n\n4. **Implementation of safeguards**: Implement technical and policy safeguards where necessary to minimize the risks of AI systems, and deploy systems after a full risk assessment.\n\n5. **Collaboration with multiple stakeholders**: Work with academia, industry, and other stakeholders to build and share culturally relevant datasets to ensure the safe and effective use of AI in localized contexts.\n\nCollectively, these methods will lead to safer and more ethical use of AI."
    }
  ]
}'

Response stream

Server-Sent Events (SSE) provides real-time streaming responses. Each SSE message consists of the following format. (Events are separated by blank lines.)

event: [event_type]
data: [json_data]

The Create conversation title API supports four event types: ping, token, result, and error.

• ping: Output data to prevent read timeouts while waiting before generating the answer (every second until the answer is generated).
• token: Output the token of the answer being generated in real time.
• result: Output the final answer.
• error: Output when an error occurs during answer generation.

Response headers

The following describes the response headers.

Response body

The response body differs depending on the event type.

ping, token, result events

The following describes the ping, token, and result events.

error event

The following describes the error events.

Response status codes

For information about the HTTP status codes common to all RAG APIs, see RAG response status codes.

Response example

The response example is as follows:

event: ping
data: {
    "message": {
        "role": "assistant",
        "content": "",
        "totalContent": ""
    },
    "status": "IN_PROGRESS",
    "references": [],
    "rateLimit": null,
    "metadata": {}
}

...

event: token
data: {
    "message": {
        "role": "assistant",
        "content": " Safe",
        "totalContent": "\"Safe AI"
    },
    "status": "IN_PROGRESS",
    "references": [],
    "rateLimit": null,
    "metadata": {}
}

...

event: result
data: {
    "message": {
        "role": "assistant",
        "content": "\"Safe AI utilization\"",
        "totalContent": "\"Safe AI utilization\""
    },
    "status": "COMPLETE",
    "references": [],
    "rateLimit": {
        "limitRequests": null,
        "remainingRequests": null,
        "resetRequests": null,
        "limitTokens": "60000",
        "remainingTokens": "58824",
        "resetTokens": "41s"
    },
    "metadata": {
        "chatModel": "HyperClovaX",
        "modelName": "HCX-005",
        "inputTokens": 327,
        "outputTokens": 8,
        "totalTokens": 335
    }
}