Adding an OpenAI integration

You can integrate your OpenAI account with HCL DevOps Loop so that can you connect to the providers and use Loop Genie services.

Before you begin

You must have the details of your subscription and valid credentials of the AI provider account to which you want to connect. See Details required for AI providers.

Procedure

  1. Click Settings > Integrations.
    The Integrations page is displayed.
  2. Click New Integration.
    The New Integration pane is displayed with the Overview tab.
  3. Enter a name to identify the integration in Name.
  4. Select Open AI from the Provider list.
  5. Select one of the following options in the Credential Type list:
    • API Key: Go to 6.
    • Password: Go to 7.
  6. Enter the API key of the AI provider account to authenticate the connection in API Key.
  7. Enter the details in the following fields:
    • Username
    • Password
    • Confirm Password
  8. Click Next.
    The Configure tab is displayed.
  9. Enter the URL of the AI provider in API Endpoint.
  10. Select Chat in Mode to set the mode of communication.
  11. Select an LLM to be used from the Model list.
  12. Enter an identifier that is configured in your OpenAI account for your organization in Organization ID.
  13. Enter an identifier that is configured in your OpenAI account for your project in Project ID.
  14. Enter the string to be used to stop the generation of text by AI in Stop Sequences.
    For example, you can set "stop" or "end" as the stop sequence.
  15. Drag the slider to set a value that determines the accuracy of the output in Temperature.
    For example, you can use low temperature for the summarization, data extraction, or details that need precision, and use higher temperature for more generic or creative responses.
  16. Drag the slider to set a value that determines the maximum length of each conversation in Max Length.
  17. Drag the slider to set a value that determines the diversity of output in Top P.
  18. Drag the slider to set a value to avoid repetition of tokens in the output in Frequency Penalty.
    Setting a higher value results in less repetitive text in responses.
  19. Drag the slider to set a value to avoid repetition of tokens that have already been used in Presence Penalty.
  20. Drag the slider to set a value that determines the number of responses to be generated in Best Of.
  21. Click Save.

Results

You have integrated your OpenAI account with DevOps Loop.