Azure AI Fundamentals

Generative Models: How They Work and Deployment

12 free questions here, all 60 in the app

Easy 20 8 of them here Medium 20 2 of them here Hard 20 2 of them here

Easy

8 of 20 questions

How generative AI models work

  1. How does a large language model (LLM) generate its response?
  2. What is a token in a language model?
  3. What does a model's context window describe?
  4. Which architecture are most of today's large language models based on?
  5. What is an embedding?
  6. What does fine-tuning a model mean?

Model selection

  1. Where in Microsoft Foundry do you find models from Microsoft, OpenAI, Meta and other providers?
  2. What characterizes a multimodal model?

Practise all 20 Easy questions in the app →

Medium

2 of 20 questions

Deployment and parameters

  1. A company needs predictable throughput with low variation in response time at consistently high volume. Which deployment type fits?
  2. A German company requires prompts and responses to be processed only within the EU but wants to pay per token. Which type fits?

Practise all 20 Medium questions in the app →

Hard

2 of 20 questions

Deployment and parameters

  1. Classify this: a retailer analyzes 2 million product descriptions overnight that must be finished by the next evening. The data must not leave the EU. Which type fits best?
  2. Which statements about deployment types in Foundry are correct?

Practise all 20 Hard questions in the app →

Certifications & exams

Which exam comes next?

You can already practise for 51 exams, and 16 are up for a vote. See all certifications with their official information, cast your vote or suggest your exam.

See all certifications