A Generative AI Engineer has fine-tuned LLMs in their production Databricks workspace and wants to make them available for testing in their development workspace. All workspaces are enabled for Unity Catalog, and models are currently being logged to the MLflow Model Registry.

What is the most cost-effective and secure method for the engineer to achieve this?

Exam-Like

Use an external model registry which can be accessed from all workspaces.

7.3%

Use MLflow to log the model directly into Unity Catalog, and enable READ access in the dev workspace to the model.

83.6%

Setup a duplicate training pipeline in dev, so that an identical model is available in dev.

5.5%

Setup a script to export the model from prod and import it to dev.

3.6%

Databricks Certified Generative AI Engineer - Associate