Cloudera Docs

Installing RAG studio

Learn about installing RAG Studio in an Air-gapped environment.

In an air-gapped environment, the RAG Studio ML Runtime image is available for download from the Cloudera Docker Registry and can be hosted in your private Docker registry.

  1. In the Cloudera console, click the Cloudera AI tile. The Cloudera AI home page is displayed.
  2. In the left navigation pane, click Cloudera AI Workbench. The Cloudera AI Workbench home page is displayed.
  3. Click on the name of the workbench. The workbench Home page is displayed.
  4. Click Site Administration in the left navigation pane.
  5. Set up the RAG Studio ML Runtime image by configuring it in the Runtimes tab for the project.
  6. [Optional] If a PyPI mirror repository is utilized, then the following environment variables must be configured under Runtimes > Environment Variables :
    1. Set UV_DEFAULT_INDEX to the PyPi mirror URL, for example, http://ccycloud.xyz-airgap.root.comops.site:8080/web/simple.
    2. Set UV_INSECURE_HOST to the PyPi mirror host, for example, ccycloud.xyz-airgap.root.comops.site.
  7. To create the RAG Studio application, use the configured ML Runtime image and the script file /rag-studio/scripts/startup_app.py to establish an application. Name this application as RagStudio.