mediumMultiple Choice
PDE Practice Question: A team trained a model on a Vertex AI custom…
A team trained a model on a Vertex AI custom training job and wants to deploy it to an endpoint for online predictions. They have the model artifacts stored in Cloud Storage. What steps are required?
⚠ Common exam trap
Google Cloud often tests the misconception that you can deploy directly from Cloud Storage without the Model Registry, or that the endpoint must be created before the model is uploaded, when in fact the model must be registered first.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Upload model to Model Registry, create endpoint, deploy model
To deploy a model for online predictions on Vertex AI, you must first upload the model artifacts from Cloud Storage to the Model Registry, which creates a versioned model resource. Then you create an endpoint (or use an existing one) and deploy the model to that endpoint, specifying machine type, traffic split, and other settings. This three-step process (upload → create endpoint → deploy) is the required workflow for online serving.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Upload model to Model Registry, create endpoint, deploy model
Why this is correct
Uploading the Cloud Storage artifacts to Vertex AI Model Registry creates the model resource, then creating an endpoint and deploying that model to it exposes it for online prediction. These three steps are the minimum sequence required.
- ✗
Directly deploy from Cloud Storage without Model Registry
Why it's wrong here
Vertex AI requires a Model resource registered before deployment; an endpoint cannot consume Cloud Storage artefacts directly. Uploading to Model Registry is genuinely needed, but as a prerequisite step, not something skipped. The stem asks for the full sequence: upload model, create endpoint, then deploy.
- ✗
Create endpoint, then upload model
Why it's wrong here
An endpoint must reference a deployed Model; creating it first leaves nothing to deploy, and Vertex AI orders these operations as upload model, create endpoint, then deploy model to endpoint. Endpoint-first suits nothing here. The stem's Cloud Storage artefacts still require a Model resource before any endpoint exists.
- ✗
Use Vertex AI Batch Prediction only
Why it's wrong here
Batch Prediction writes outputs to Cloud Storage and creates no endpoint, so it cannot serve online requests. It is the right choice for large asynchronous scoring jobs where latency is irrelevant. The stem explicitly requires an endpoint for online predictions, which batch prediction never provisions.
Go deeper
Related to this question
About these practice questions
Courseiva writes every PDE question from scratch — 747 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PDE practice question is part of Courseiva's free Google Cloud certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the PDE exam.