AI-900 Practice Question: Describe features of computer vision workloads on Azure
What is the difference between Azure AI Vision and Azure AI Custom Vision in terms of when to use each?
⚠ Common exam trap
Test-takers frequently confuse 'general vs. specialized' with superficial attributes like image size or source, leading them to pick options that sound plausible but miss the core functional difference between pre-trained and custom-trained models.
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 Azure AI Vision for general image analysis; use Custom Vision when you need specialized domain-specific recognition
Azure AI Vision is a pre-trained service for general image analysis tasks like object detection, OCR, and description generation, requiring no custom training. Azure AI Custom Vision allows you to train a model on your own labeled images for specialized, domain-specific recognition tasks, such as identifying unique product defects or rare animal species. Option B correctly captures this distinction: use Azure AI Vision for broad, out-of-the-box capabilities and Custom Vision when you need tailored recognition for your specific use case.
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 Azure AI Vision for large images; use Custom Vision for small images
Why it's wrong here
Image dimensions or file size do not separate these services; Azure AI Vision imposes size limits (for example, 4 MB for many features), and Custom Vision also expects images under a certain size (typically 6 MB for training and 4 MB for prediction), so very large images in either service would need resizing. The wrong assumption focuses on a technical constraint rather than the intended functionality. The correct basis is always general pre-built AI versus custom model training, not image resolution.
- ✓
Use Azure AI Vision for general image analysis; use Custom Vision when you need specialized domain-specific recognition
Why this is correct
Azure AI Vision is the right choice when you need immediate, pre-trained analysis of everyday images—identifying objects, reading printed or handwritten text, generating captions, or checking for offensive content. Custom Vision is required when your classification targets are specialized or unique, such as detecting defective parts in a factory, distinguishing plant diseases, or recognizing specific animal breeds, because Azure AI Vision does not contain these categories. In short, choose Vision for general inspection and Custom Vision for purpose-built domain recognition.
- ✗
Use Azure AI Vision only in production; Custom Vision only in development
Why it's wrong here
The deployment stage does not differentiate the two services: Azure AI Vision is a fully managed REST API suitable for production, and Custom Vision can be exported as a TensorFlow/ONNX model or Docker container and deployed in production environments. Using Custom Vision in development only ignores its production-ready export options and scalability. The real basis is the scope of recognition—pre-built general capabilities versus custom-trained domain-specific models.
- ✗
Use Azure AI Vision for images from cameras; Custom Vision for images from documents
Why it's wrong here
The source of the image is irrelevant to the service choice: Azure AI Vision can analyze camera snapshots, and Custom Vision can accept document images for custom classification. Azure AI Vision includes OCR capabilities specifically for extracting text from documents, while Custom Vision performs classification or object detection on whatever images you upload. The actual decision is whether you need pre-built, general-purpose recognition or a model trained on your own domain-specific labels.
Go deeper
Related to this question
Learn chapter
Azure Machine Learning Studio
Key term
Model
In IT and AI, a model is a trained mathematical representation that learns patterns from data to make predictions or decisions.
Key term
Service
A service is a software component or system that performs a specific function and is available to be used by other programs or users over a network.
About these practice questions
Courseiva writes every AI-900 question from scratch — 985 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 AI-900 practice question is part of Courseiva's free Microsoft 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 AI-900 exam.