Courseiva

AI-900 Practice Question: Describe features of generative AI workloads on Azure

What is 'Azure OpenAI deployment' and how does it differ from a 'model'?

⚠ Common exam trap

Many candidates confuse the conceptual 'model' (the AI algorithm) with the operational 'deployment' (the provisioned instance), often assuming they are interchangeable or that a deployment is merely a 'copy' of the model, missing the critical quota and endpoint management aspects.

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

A model is the underlying AI; a deployment is a named, quota-allocated instance your application calls

In Azure OpenAI, a 'model' refers to the underlying AI algorithm (e.g., GPT-4, GPT-3.5-Turbo) that defines the capabilities and behavior of the generative AI. A 'deployment' is a specific, named instance of that model provisioned within an Azure OpenAI resource, with its own endpoint, quota (tokens per minute), and configuration (e.g., content filter settings). This separation allows you to manage capacity and access for different applications or use cases independently, even when using the same base model.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

  • A model is the purchased licence; a deployment is the technical installation

    Why it's wrong here

    Azure OpenAI does not sell model licenses; it offers models as consumption-based managed APIs. A deployment is not an installation but a named model instance with its own endpoint, quota, and content-filter settings. You pay for tokens processed by that deployment, never for the model itself.

  • A model is the underlying AI; a deployment is a named, quota-allocated instance your application calls

    Why this is correct

    The model is the pre-trained neural network, such as GPT-4o, that provides the core inference capability. When you create a deployment, you instantiate that model under a logical name, assign a throughput quota (tokens per minute), and obtain a REST endpoint plus API key. You can create multiple deployments of the same model, e.g., 'dev' and 'prod', to isolate workloads and manage rate limits independently.

  • A deployment is always faster than a model because it uses optimised serving infrastructure

    Why it's wrong here

    A deployment does not change the model's latency or algorithmic speed; any performance difference comes from the underlying model and the Azure infrastructure serving it. Increasing a deployment's quota raises the allowed token-throughput ceiling, but that is capacity, not optimisation. Even a low-quota deployment runs the same model inference operations.

  • Models are available globally; deployments are restricted to specific Azure regions

    Why it's wrong here

    Regional availability applies equally to models and deployments: a model is available only in the Azure regions where Microsoft hosts it, and each deployment is created in a specific region. Deployments are not arbitrarily restricted beyond that; they are endpoints tied to the chosen region's infrastructure. Calling a model 'global' is imprecise because both the model catalogue and every deployment are regionally scoped.

About these practice questions

This AI-900 question is part of Courseiva's 985-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →

How Courseiva writes practice questions · Editorial policy

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.