Use Azure OpenAI in Foundry Models to Generate Content AI-102 Practice Questions (Page 1)

Questions 1–5

1. 1application · easy

   A developer has provisioned an Azure OpenAI resource and wants to deploy a model for code generation. The developer needs to choose a deployment name and model version. What is the correct next step in Foundry Models?

   • ACreate a new resource group and then deploy the model
   • BGenerate an API key before deploying the model
   • CWrite application code before deploying the model
   • DSelect the desired model, choose a deployment name, and create the deployment
   Reveal Solution Discussion Select an answer first
2. 2expert · hard

   A company is building a real-time translation service using Azure OpenAI. The service must handle high throughput and low latency. The team has deployed gpt-4o-mini. However, they notice that some translations are inaccurate for technical terms. What should they do to improve accuracy without significantly increasing latency?

   • ASwitch to gpt-4o for all requests
   • BUse a system prompt with a glossary of technical terms
   • CIncrease the max_tokens parameter
   • DUse a lower temperature setting
   Reveal Solution Discussion Select an answer first
4. 3expert · hard

   A company is building a customer support chatbot that must handle both natural language conversations and generate code snippets for technical users. They have a limited budget and need to minimize latency. They currently have no Azure OpenAI deployments. What should they do?

   • ADeploy a single gpt-4o-mini model and use it for both chat and code generation
   • BDeploy gpt-35-turbo for chat and text-embedding-3-large for code generation
   • CDeploy two separate gpt-4o models, one for chat and one for code generation
   • DDeploy gpt-4o for chat and DALL-E 3 for code generation
   Reveal Solution Discussion Select an answer first
5. 4application · easy

   A user wants to use Azure OpenAI to draft a professional email. The user has access to Foundry Models and needs to select a model. Which model should the user choose for this natural language generation task?

   • Agpt-4o
   • Bwhisper
   • Ctext-embedding-3-small
   • DDALL-E 3
   Reveal Solution Discussion Select an answer first
6. 5expert · hard

   A developer is using Azure OpenAI to generate code for a complex algorithm. The model's responses are sometimes incorrect or incomplete. The developer wants to improve the quality of the generated code without changing the model. What should they do?

   • AUse the text-embedding-3-large model to generate code
   • BSwitch to a DALL-E 3 deployment for code generation
   • CRefine the prompt to include more specific requirements, constraints, and examples
   • DIncrease the temperature to 2.0 to get more deterministic outputs
   Reveal Solution Discussion Select an answer first