Create a completion profile - Fluid Topics - Latest

Use completion profiles for non dialog-based prompts. Completion profiles process content using a single prompt and produce a single output.

To create a Completion profile, go through the following steps:

1. Select New profile.

2. Select Completion.

3. Enter or select:

   • A name.

     The profile name and ID must be different from existing profiles.

   • A unique profile ID.

   • A provider: choose one of Amazon Bedrock, Anthropic, Azure, Fluid Topics, Google, MyLLM, or OpenAI.

     The MyLLM option makes it possible to provide the URL to an LLM deployment, making it possible to integrate self-hosted models with Fluid Topics. The LLM server API must be compliant with the OpenAI API specification.

   • An LLM model from the selected provider.

     • A host and model name for Azure and MyLLM.

     The LLM model dropdown menu is not exhaustive. It is possible to enter any value, as long as it matches a valid model from the selected LLM provider. Use the Test connection button to check.

   • An API key from the provider, or OAuth 2.0 client credentials.

   • A prompt.

     A prompt is a natural language request submitted to a language model to receive a response back. Source: Introduction to prompting

     The prompt can contain variables. Place variables in single curly brackets. Users can set the values of these variables when using the Query a large language model (LLM) web service.

4. Select Save.

Once created, ADMIN users can copy the completion profile ID to use in AI features.

Completion AI features include:

• The AI Case deflection component
• The AI JavaScript method for Custom components