LLM Manager

The LLM Manager allows admins to create and govern LLM Providers. The providers are used to generate images, scripts, excel formulas, translations, and to choose colors using AI. Providers can be added, removed, and edited.

LLM Provider Manager

The LLM Provider Manager lists the current details for the providers:

  • Name: The provider name
  • Type: Open AI and Azure AI.
  • Endpoint: the endpoint defined for the LLM provider.

For each provider administrators can:

  • Edit: opens and edits the given provider in the bottom panel.
  • Delete: deletes the selected provider.
  • Copy: copy the provider and allow the admin to provide a name for the newly created (copied) provider.
  • Audit Trail: display an audit trail of the selected provider in a pop-up window, detailing, operation, performed by, date, and details.

Add an LLM Provider

  • To add an LLM provider, click Add LLM Provider.
  • To edit an existing LLM Provider, click the row representing the LLM Provider that you want to enter in the top panel.

In both cases, the LLM Settings panel opens at the bottom of the page.

LLM Settings

Note: Some vendors are still in Beta at the time of writing and may not be visible in GA

  • Vendor: Select the required vendor from the drop-down list. This is typically one of OpenAI or Azure AI. The following additional options are Beta Release only: Google Vertex, IBM Watson, AI21.
  • Name: provide a name for the LLM provider.
  • OpenAI Options: If OpenAI is selected then enter the following details:
    • API Key: enter your API key, which can be found in the Account Settings of your OpenAI account.
  • Azure AI Options: If Azure AI is selected then enter the following details: 
    • API Key: enter your API key, which can be found in your Azure portal in the Endpoint and Keys pane of the Resource Management section.
    • Endpoint: enter your Azure Endpoint URL, which can be found in your Azure portal in the Endpoint and Keys pane of the Resource Management section.
    • Deployment: enter the name of the relevant Azure Deployment Environment.
  • Google Vertex Beta Options: If Google Vertex is selected then enter the following details: 
    • Region: select the region providing the Google Vertex service from the drop-down.
    • Project Id: enter the project ID for the provider
    • Model Id: enter the model ID for the provider
    • Client Id: enter the client ID for the provider
    • Client Secret: enter the client secret for the provider
    • Click on the Sign-In with Google button to test the service
    • OAuth refresh code: enter the OAuth refresh code required .
  • IBM Watson Beta Options: If IBM Watson is selected then enter the following details: 
    • API Key: enter your API key.
    • Region: select the region providing the service from the drop-down.
    • Project Id: enter the project ID for the provider
    • Model Id: enter the model ID for the provider
  • AI21 Beta Options: If AI21 is selected then enter the following details: 
    • API Key: enter your API key.
    • Model Id: enter the model ID for the provider

LLM Settings - Buttons

  • Click Test to test the LLM provider details in the panel. Note: This test will fail if the main Generative AI and Natural Language Querying option is not enabled at the top of the AI Settings panel.
  • Click Save or Apply to save your changes. Save is offered when you are creating a new LLM provider, and Apply when you are editing details of an existing LLM Provider.