Practice free →
HomeGCP ML Engineercloudcomputingmleops › Which GCP service should an ML Engineer use to D…

Which GCP service should an ML Engineer use to DEPLOY a fine-tuned LLM as a serverless container endpoint — with traffic-based autoscaling and per-request billing?

ACloud DNS
BCloud Run (with a model-serving container) — or a Vertex AI endpoint for full ML platform integration
CCloud Memorystore
DLong-lived VM with manual scaling
Answer & Solution
Correct answer: B. Cloud Run (with a model-serving container) — or a Vertex AI endpoint for full ML platform integration
Cloud Run is GCP's serverless-container runtime (per PMLE §5.1). The other options aren't serverless.
Solve this in the app — GCP ML Engineer practice & 24k+ MCQs →
Related questions