hardMultiple Choice
PMLE Practice Question: A data science team is deploying a PyTorch model…
A data science team is deploying a PyTorch model for real-time inference using Vertex AI Endpoints. The model requires a custom container with specific CUDA drivers and Python packages. They have created a Docker image and pushed it to Artifact Registry. The pipeline should automatically retrain the model every week and deploy the new version if it passes validation. However, the deployment step fails intermittently with the error 'The container image is not compatible with the machine type.' What is the most likely cause?
⚠ Common exam trap
Google Cloud often tests the distinction between deployment-time compatibility errors and runtime health check failures, tricking candidates into confusing a misconfigured machine type with a failing health probe.
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
✓
The container image requires GPU support but the machine type specified in the endpoint is a CPU-only machine.
The error 'The container image is not compatible with the machine type' indicates a mismatch between the container's hardware requirements and the machine type selected for the Vertex AI Endpoint. Since the custom container requires specific CUDA drivers, it is built for GPU acceleration. If the endpoint is configured with a CPU-only machine type (e.g., n1-standard-4), the container will fail to run because the GPU drivers cannot initialize, triggering this incompatibility error.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
The service account does not have permission to pull the container from Artifact Registry.
Why it's wrong here
Artifact Registry permission failures surface as image-pull errors, not machine-type incompatibility, which Vertex AI raises when the container's CUDA or architecture requirements exceed the selected machine's GPU or CPU. Granting the service account Artifact Registry Reader is the fix when pulls are denied, but that addresses authorisation, not hardware compatibility.
- ✓
The container image requires GPU support but the machine type specified in the endpoint is a CPU-only machine.
Why this is correct
Vertex AI validates the container against the endpoint's machine type; a CUDA-dependent image cannot start on a CPU-only node, producing the incompatibility error. Selecting a GPU-backed machine type aligns the runtime environment with the image's driver requirements.
- ✗
The container's health check endpoint is not responding correctly.
Why it's wrong here
A failing health check produces readiness or liveness probe errors after the container starts, not a machine-type compatibility message during deployment. It is tempting because misconfigured probes genuinely break Vertex AI deployments, but that would be correct if the error reported unhealthy or unresponsive container endpoints rather than incompatible images.
- ✗
The model artifact size exceeds the maximum allowed for the machine type.
Why it's wrong here
Artifact size does not trigger a machine-type compatibility error; that message arises from mismatched architecture, CUDA or driver requirements between image and selected machine. It is tempting because oversized models do fail deployment, but that would be correct if the error cited memory or storage limits instead of container compatibility.
Go deeper
Related to this question
About these practice questions
One of 775 original PMLE practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PMLE 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 PMLE exam.