Home / ML Associate practice test / Databricks Machine Learning / Question

Databricks Machine Learning Associate · free practice question

A production registered model in Unity Catalog already has version 3 marked as Challenger after validation. Version 2 currently serves as Champion. The new…

Exam-style question on Databricks Machine Learning from the free ML Associate practice test, with the answer and a full explanation below.

Updated November 10, 2026

Databricks Machine Learning

A production registered model in Unity Catalog already has version 3 marked as Challenger after validation. Version 2 currently serves as Champion. The new version is approved for deployment, and the team wants downstream jobs that reference Champion to pick up the new version without changing their code. What should they do?

  1. ARename the registered model itself from fraud_model to fraud_model_v3, because batch jobs and endpoints always follow the newest model name rather than a mutable alias reference
  2. BCopy version 3 into a second registered model called Champion, because alias changes do not affect existing inference jobs and only a new registered model name can trigger deployment updates
  3. CDelete version 2 first, because Unity Catalog requires the previous Champion version to be removed before another version in the same model can take over production traffic
  4. DCreate a new workspace experiment named Champion and move version 3 there, because Databricks serving and batch inference resolve deployment state from experiment names instead of registry aliases
  5. EReassign the Champion alias from version 2 to version 3 using the MLflow Client API so alias-based consumers automatically resolve the newly approved production version
Answer & explanation

Correct answer: E

WHY E: Databricks recommends using aliases such as Champion and Challenger; reassigning Champion updates alias-based consumers without changing their model reference. WHY NOT A: Renaming the model is unnecessary and more disruptive. WHY NOT B: A second model is not required. WHY NOT C: The old version does not need to be deleted first. WHY NOT D: Experiments do not define deployment aliases.

Take the full ML Associate practice test free →

Would you pass today?

This question is one of 45 in the free, timed Databricks Machine Learning Associate practice test. Take it and get a section-by-section score.

Start the free test →