To configure this service:

1. Choose Settings > Configuration.
2. In the Large Language Model Services section of the page, click + New LLM Service.
3. Complete the row:
   • Name: Enter an easily identifiable name for the LLM service. This is the name that you'll choose when you configure the dialog flow.
   • LLM Service: Select from the LLM services that have been configured for the instance.
   • Transformation Handler: Choose the transformation handler that's the counterpart of the LLM service. For example, if the LLM component uses the Azure OpenAI LLM service to generate responses, then you would choose one of the skill's event handlers that transforms the Azure OpenAI payloads to CLMI.
   • Mock: Switch this option on (true) to save time and costs when you test your dialog flow. Because this option enables the LLM component to return a static response instead of the LLM-generated continuation, you don't have to waste time waiting for the LLM response, nor do you incur costs from the LLM service provider when you're just testing out part of the dialog flow.
     Note
     
     You can only use this option if the LLM service that you've selected has a 200 static response.
   • Default: Switching this option on (true) sets the LLM service as the default selection in the LLM component's LLM Service menu. If an existing LLM Service is already set as default, then its default status gets overwritten (that is, set to false) when you switch this on for another LLM Service.
4. Click the Save action (located at the right).
   
   
   
   

LLM services created as REST services in releases prior to 24.02 are flagged with a warning . To dismiss this warning, select the REST service (accessed through Settings > API Services), then click Convert to LLM.