A company wants to build an application that can analyze both text and images to answer user questions. They need to support a wide range of input types, including documents with embedded images. They have a limited budget and want to minimize the number of deployments. What should they do?

Deploy gpt-35-turbo for text and DALL-E 3 for images

Deploy text-embedding-3-large for both text and image analysis

Deploy a single gpt-4o model that accepts both text and image inputs

Deploy two separate gpt-4o models, one for text and one for images

What is the input required for DALL-E 3 to generate an image?

A text prompt describing the desired image

A team has provisioned an Azure OpenAI resource and deployed a gpt-4o model. They are now building an application that needs to call the model. They want to ensure the application can access the deployment securely without exposing the API key. What should they do?

Embed the API key directly in the application's source code

Store the API key in a public GitHub repository

Use the deployment name as the authentication secret

Use Azure Active Directory (Azure AD) authentication with a managed identity for the application

Microsoft Certified:Azure AI Engineer Associate

Domain 2Objective 2

Use Azure OpenAI in Foundry Models to Generate Content AI-102 Practice Questions (Page 6)

Part of the Implement generative AI solutions domain, which accounts for 15-20% of the AI-102 exam. Microsoft does not publish an official question count, but from its 100-minute exam (~40–65 total, ~6–13 in this domain), expect 2–4 from this objective — we provide 49 practice questions to prepare you well beyond it. (estimate)

49questions here

10free pages

8concepts

15-20%of the exam

Questions 26–30

26foundation · easy
Which Azure OpenAI model is used to generate images from textual descriptions?
Select an answer first
27expert · hard
A company wants to build an application that can analyze both text and images to answer user questions. They need to support a wide range of input types, including documents with embedded images. They have a limited budget and want to minimize the number of deployments. What should they do?
Select an answer first
28foundation · easy
What is the input required for DALL-E 3 to generate an image?
Select an answer first
29expert · hard
A team has provisioned an Azure OpenAI resource and deployed a gpt-4o model. They are now building an application that needs to call the model. They want to ensure the application can access the deployment securely without exposing the API key. What should they do?
Select an answer first
30application · medium
A developer needs to generate Python code snippets that convert JSON data into CSV format. The team has an Azure OpenAI resource in Foundry Models with several model deployments. The developer wants to minimize token usage while ensuring the model can reliably follow complex formatting instructions. Which model deployment should the developer use?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to AI-102

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Microsoft. “AI-102” is a trademark of its owner, used for identification only.

Use Azure OpenAI in Foundry Models to Generate Content AI-102 Practice Questions (Page 6)

Questions 26–30

Which Azure OpenAI model is used to generate images from textual descriptions?

A company wants to build an application that can analyze both text and images to answer user questions. They need to support a wide range of input types, including documents with embedded images. They have a limited budget and want to minimize the number of deployments. What should they do?

What is the input required for DALL-E 3 to generate an image?

A team has provisioned an Azure OpenAI resource and deployed a gpt-4o model. They are now building an application that needs to call the model. They want to ensure the application can access the deployment securely without exposing the API key. What should they do?

Keep studying — Implement generative AI solutions