Importing models from NVIDIA NGC

You can import the NVIDIA NGC Catalog models listed in the Model Hub page and deploy it from the Registered Models page.

To import models, you must add the following NVIDIA GPU Cloud (NGC) URL details so they can be allowed in the firewall's rules.
  • prod.otel.kaizen.nvidia.com (NVIDIA open telemetry)
  • api.ngc.nvidia.com
  • files.ngc.nvidia.com
  1. In the Cloudera Data Platform console, click the Cloudera AI tile.
    The Cloudera AI Workbenches page displays.
  2. Click Model Hub under AI Hub in the left navigation menu.
    The Model Hub page displays. The page lists different models along with its source type, tags, and description.
  3. Click Import on the model you want to import.
    The Import Model page displays.
  4. In the Select Registry drop-down list, select the AI registry to which you want to import the model.
  5. In the Select Parameter drop-down list, select the model variant.
  6. In the Select Optimization drop-down list, select the model variant that is optimized for your use case and domain.
    The details of the chosen optimization profile is displayed.
  7. In the Enter Name or Select Model field, select a name from the existing list or enter a new name for the model you are importing.
  8. If displayed, read the User Policy and License Agreement, and click the checkbox if you agree.
  9. Click Import.
    The Model Hub page is displayed with the message that the model is imported successfully.
You can click Registered Models in the left navigation menu to view the newly imported model.