AI Assistant settings

Note

This feature is available in Enterprise, AWS, and Team editions only.

Use this page to configure the general behavior of AI in CloudBeaver.

AI Assistant setup

To activate the AI features in CloudBeaver, configure the API token:

1. Navigate to Settings -> Administration -> Server Configuration
2. Ensure the AI option is activated
3. Navigate to AI Settings tab

4. Use the Engine dropdown menu to switch between and set up supported providers:

   • OpenAI
   • GitHub Copilot
   • Azure OpenAI
   • Google Gemini
   • Ollama
   • Anthropic Claude
   • Amazon Bedrock

Send only relevant objects to AI

Available since 25.3

This option enables Retrieval-Augmented Generation behavior (RAG). When it’s on, CloudBeaver looks at your request and sends only the tables, columns, or other objects that are actually needed for the answer. This keeps responses faster, reduces token usage, and helps the assistant build replies based on real, relevant data instead of your full schema.

Note

RAG works only with the OpenAI provider, and only when the server runs on a PostgreSQL database. For information on setting up the server database, see Server database.

Disable AI Assistant

To disable AI Assistant features:

1. Navigate to Settings -> Administration -> Server Configuration -> Services section
2. Deselect AI option
3. Save changes