Question 338 of 1,672
MLS-C01 Modeling Practice Question
A data science team is deploying a machine learning model to production using SageMaker. The model is a PyTorch model that requires custom inference logic including image preprocessing. The team needs to ensure that the endpoint can handle variable batch sizes and has low latency. Which deployment approach should the team use?
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
✓
Use SageMaker Inference Pipelines with a preprocessing container followed by the PyTorch model container.
SageMaker Inference Pipelines allow chaining of preprocessing and prediction containers, which enables custom inference logic like image preprocessing and supports variable batch sizes with low latency. Option B (AWS Lambda with API Gateway) is not designed for real-time inference with large models and variable batch sizes due to execution time and payload limits. Option C (SageMaker Python SDK's Predictor class) is a client interface to invoke an already deployed endpoint, not a deployment approach. Option D (SageMaker multi-model endpoint) hosts multiple models on a single endpoint but does not directly address custom inference logic or preprocessing. Option E (SageMaker Batch Transform) is for offline batch inference, not real-time low-latency inference.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Use SageMaker Inference Pipelines with a preprocessing container followed by the PyTorch model container.
Why this is correct
Inference Pipelines allow custom preprocessing and model serving with low latency.
- ✗
Deploy the model as an AWS Lambda function and use API Gateway.
Why it's wrong here
Lambda has time and memory limits, unsuitable for large model inference.
- ✗
Use the SageMaker Python SDK's Predictor class with the model artifact.
Why it's wrong here
The Predictor is a client object, not a deployment infrastructure.
- ✗
Use a SageMaker multi-model endpoint to host the model with a custom container.
Why it's wrong here
Multi-model endpoints are for hosting multiple models, not specifically for custom preprocessing logic.
- ✗
Use SageMaker Batch Transform for real-time inference.
Why it's wrong here
Batch Transform is for asynchronous batch processing, not real-time.
Quick reference
Cloud Service Model Comparison
| Model | You Manage | Provider Manages | Examples |
|---|---|---|---|
| IaaS | OS, runtime, apps, data | Hardware, hypervisor, networking | EC2, Azure VMs, GCP Compute Engine |
| PaaS | Apps and data | OS, runtime, middleware, hardware | Elastic Beanstalk, Azure App Service |
| SaaS | Data and settings only | Everything else | Microsoft 365, Salesforce, Workday |
| FaaS / Serverless | Function code only | Infra, scaling, runtime | Lambda, Azure Functions, Cloud Run |
| CaaS | Containers and apps | Kubernetes, OS, hardware | EKS, AKS, GKE |
About these practice questions
Courseiva creates original exam-style practice questions with explanations and wrong-answer analysis. It does not publish real exam questions, exam dumps, or protected exam content. Learn why practice questions differ from exam dumps →
Last reviewed: Jun 20, 2026
This MLS-C01 practice question is part of Courseiva's free Amazon Web Services 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 MLS-C01 exam.
Question Discussion
Share a tip, memory trick, or ask about the reasoning behind this question. Do not post real exam questions, leaked content, braindumps, or copyrighted exam material. Comments are moderated and may be removed without notice.
Sign in to join the discussion.