Deploying BigQuery ML Models via Vertex AI Model Registry

PMLE By islever Editorial Last updated

Serving models. Automating model retraining.

Short answer

C Alter the model by using BigQuery ML to register it with Vertex AI Model Registry, then deploy from there to a Vertex AI endpoint.

Question

You recently used BigQuery ML to train an AutoML regression model. You shared results with your team and received positive feedback. You need to deploy your model for online prediction as quickly as possible. What should you do?
  1. Retrain the model by using BigQuery ML, and specify Vertex AI as the model registry. Deploy the model from Vertex AI Model Registry to a Vertex AI endpoint,
  2. Retrain the model by using Vertex Al Deploy the model from Vertex AI Model. Registry to a Vertex AI endpoint.
  3. Alter the model by using BigQuery ML, and specify Vertex AI as the model registry. Deploy the model from Vertex AI Model Registry to a Vertex AI endpoint. Correct Answer
  4. Export the model from BigQuery ML to Cloud Storage. Import the model into Vertex AI Model Registry. Deploy the model to a Vertex AI endpoint.

Key Takeaways

Use ALTER MODEL to register BigQuery ML models with Vertex AI Model Registry for faster deployment compared to exporting and importing.

Learn how to quickly deploy an existing BigQuery ML AutoML regression model for online prediction by registering it with Vertex AI Model Registry using ALTER MODEL.

Common mistake: Users often choose Option D (Export/Import) because they assume models must be physically moved to Vertex AI. However, this is slower than direct registration via ALTER MODEL.

Expert Analysis

The question requires deploying an existing BigQuery ML AutoML regression model for online prediction as quickly as possible. The key constraint is speed and avoiding unnecessary steps like retraining or manual export.

Why Option C is Correct: BigQuery ML supports registering models directly with the Vertex AI Model Registry using the ALTER MODEL statement. This action links the BigQuery ML model to the Vertex AI ecosystem without moving data or requiring a new training job. Once registered, the model can be deployed to a Vertex AI endpoint for online serving. This is the most efficient path because it leverages the existing trained model and uses a direct registration mechanism.

Why Other Options Are Incorrect:

  • Options A and B (Retraining): The prompt states the team gave "positive feedback," implying the model is ready. Retraining adds significant time and cost, violating the "as quickly as possible" requirement. Additionally, Option B suggests retraining in Vertex AI, which ignores the work already done in BigQuery ML.
  • Option D (Export/Import): While exporting to Cloud Storage and importing into Vertex AI is a valid workflow, it involves multiple steps (export, upload, import) and is slower than the direct registration provided by ALTER MODEL. It is not the quickest method.

Community comments largely support Option C, highlighting that ALTER MODEL allows registration without export, making it faster than Option D's export/import process. Some users incorrectly suggest Option B due to misconceptions about compatibility, but BigQuery ML models are compatible with Vertex AI serving.

Official Reference

Exam Strategy

When a question emphasizes 'quickly' or 'fastest' deployment of an existing model, look for options that avoid retraining or manual data movement. In Google Cloud ML, check if services like BigQuery ML offer direct integration points (like Model Registry registration) rather than generic export/import paths. Eliminate retraining options immediately if the model is already validated.

Frequently Asked Questions

Can I deploy BigQuery ML models directly to Vertex AI endpoints?

Yes, you can register BigQuery ML models with Vertex AI Model Registry and then deploy them to Vertex AI endpoints for online prediction.

Is it necessary to export BigQuery ML models to Cloud Storage before deploying?

No, you can use the ALTER MODEL statement to register the model with Vertex AI Model Registry without exporting it to Cloud Storage first.

More PMLE FAQ →

Community Votes

C
71%
D
23%
B
6%

71% of anonymous learners picked answer C. Votes are pick records left by other test-takers — they are not the verified answer.

Discussion 28 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 28

  1. Dagogi96 👍 8
    Friend is the C, and with Alter MODEL you can register the model in Vertex AI, I work in a company and I myself have registered models like this.
  2. cruise93 👍 8 Answered C
    https://cloud.google.com/bigquery/docs/reference/standard-sql/bigqueryml-syntax-alter-model#alter_model_statement
  3. pikachu007 👍 7 Answered D
    I think it's D, as model retraining should not be required unless it's specified there's new data.
  4. vaibavi 👍 5
    I think it's C Exported models for model types AUTOML_REGRESSOR and AUTOML_CLASSIFIER do not support AI Platform deployment for online prediction.
  5. sonicclasps 👍 4 Answered C
    the answer is C, no need to export the model : "You can register BigQuery ML models with the Model Registry, in order to manage them alongside your other ML models without needing to export them" https://cloud.google.com/bigquery/docs/managing-models-vertex a simple update is sufficient : https://cloud.google.com/bigquery/docs/update_vertex
Show 23 more comments
  1. phani49 👍 4 Answered C
    You can use the ALTER MODEL statement to register your existing BigQuery ML model with Vertex AI Model Registry https://cloud.google.com/bigquery/docs/reference/standard-sql/bigqueryml-syntax-alter-model
  2. lunalongo 👍 3 Answered C
    C is the best option because: 1) Retraining the model (A/B) is not necessary (see positive feedbacks) 2) Exporting to Cloud Storage (D) is not necessary, since you can use the ALTER MODEL statement to register it on Vertex AI Model Registry and deploy it to the Vertex AI endpoint from there 3) Using BigQuery ML without exporting the model is the quickiest option
  3. gscharly 👍 3 Answered C
    https://cloud.google.com/vertex-ai/docs/model-registry/model-registry-bqml https://cloud.google.com/bigquery/docs/update_vertex
  4. bigdapper 👍 3 Answered C
    Ans: C (ALTER MODEL) https://cloud.google.com/bigquery/docs/managing-models-vertex#sql
  5. pinimichele01 👍 3
    alter the model doesn't mean retrain...
  6. fitri001 👍 2 Answered D
    You recently used BigQuery ML to train an AutoML regression model. You shared results with your team and received positive feedback. You need to deploy your model for online prediction as quickly as possible. What should you do? A. Retrain the model by using BigQuery ML, and specify Vertex AI as the model registry. Deploy the model from Vertex AI Model Registry to a Vertex AI endpoint, B. Retrain the model by using Vertex Al Deploy the model from Vertex AI Model. Registry to a Vertex AI endpoint. C. Alter the model by using BigQuery ML, and specify Vertex AI as the model registry. Deploy the model from Vertex AI Model Registry to a Vertex AI endpoint. D. Export the model from BigQuery ML to Cloud Storage. Import the model into Vertex AI Model Registry. Deploy the model to a Vertex AI endpoint.
  7. omermahgoub 👍 2
    I changed my answer to C. GCS is not necessary
  8. shadz10 👍 2
    I agree with pikachu007
  9. taylz876 👍 1 Answered C
    Why C is right You already have a trained model that your team approved. You don't need to retrain it, only make it available for online prediction. BigQuery ML can register an existing model in the Model Registry with an ALTER MODEL statement that sets the Vertex AI model ID. The model isn't moved or copied. It's added to the catalog. From the Model Registry you deploy it to an endpoint, which serves online predictions. That's the shortest path, with no retraining and no export. Why the others are wrong • A: Retraining the model in BigQuery ML wastes time and compute, and the trained model would be different from the one your team reviewed. • B: Retraining in Vertex AI is the same problem, plus it rebuilds the model in a different place. • D: Exporting to Cloud Storage and then importing into the Model Registry also works, but it's more steps. Your notes say registering doesn't require moving the model.
  10. fitri001 👍 1
    No Retraining: You've already trained a successful model in BigQuery ML. Retraining (Options A, B, and C) is unnecessary and adds time. Direct Deployment: Option D leverages existing tools for streamlined deployment. You export the model directly from BigQuery ML and import it into Vertex AI Model Registry for centralized management. Finally, you deploy the model to a Vertex AI endpoint for online predictions. Cloud Storage: Cloud Storage provides a readily accessible location to store your exported model before deployment.
  11. AzureDP900 👍 1
    C is correct Here's why: 1) You trained an AutoML regression model using BigQuery ML. 2)To deploy the model for online prediction, you need to export the model in a format that is compatible with Vertex AI. 3)Altering the model by using BigQuery ML and specifying Vertex AI as the model registry allows you to export the model in the correct format. Once exported, you can deploy the model from Vertex AI Model Registry to a Vertex AI endpoint, which enables online prediction
  12. lunalongo 👍 1
    D is the best option because: 1) BigQuery ML is excellent for training, not much for online prediction 2) Vertex AI provides a more robust and scalable infrastructure. 3) Exporting model from BigQuery ML to a format compatible with Vertex AI (typically Cloud Storage) is required *A, B, and C attempt to directly deploy from BigQuery ML, which isn't a supported workflow.
  13. alfieroy16 👍 1 Answered C
    Alter the model is correct,no need to export the model : "You can register BigQuery ML models with the Model Registry, in order to manage them alongside your other ML models without needing to export them" https://cloud.google.com/bigquery/docs/managing-models-vertex a simple update is sufficient : https://cloud.google.com/bigquery/docs/update_vertex
  14. vaibavi 👍 1 Answered B
    I think the answer here is B , because even if we alter or export automl regressor model trained in BQML is not supported in vertex ai for online prediction so we need to retrain using vertex ai
  15. omermahgoub 👍 1 Answered D
    D. Export the model from BigQuery ML to Cloud Storage. Import the model into Vertex AI Model Registry. Deploy the model to a Vertex AI endpoint.
  16. d6e1ae4 👍 1 Answered D
    The model has already been trained and received positive feedback, so there's no need to retrain the model.
  17. Land3r 👍 1 Answered C
    https://cloud.google.com/bigquery/docs/managing-models-vertex#register-new-bqml-model-version
  18. playerXL7 👍 1 Answered C
    https://cloud.google.com/vertex-ai/docs/model-registry/model-registry-bqml
  19. itwiz 👍 1
    C) https://cloud.google.com/bigquery/docs/create_vertex
  20. pinimichele01 👍 1
    why not C? it is not necessary to export in GCS
  21. Omi_04040 👍 1 Answered C
    No need to export the model to Cloud Storage
  22. daidai75 👍 1
    Agree with Pikachu007, the option D is good.
  23. hybridpro 👍 1 Answered C
    It's C

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Practice All PMLE Questions

Access 65 questions with complete answers and detailed explanations.

View Full PMLE Practice Test →

← Back to PMLE Study Guide