AI-300 Certification Practice Question #24

Question

A company is creating an internal tool that summarizes long meeting transcripts and extracts action items.
The model must:
Process text inputs up to 200k tokens long.
Generate concise summaries in seconds.
Support interactive testing before integration into the app.
You need to select, deploy, and test a model that supports summarization with low latency.
How should you complete the configuration plan? To answer, select the appropriate options in the answer area.
NOTE: Each correct selection is worth one point.

Accepted Answer

The model must accept up to **200k-token** inputs and return concise summaries with low latency — **GPT-4.1-Mini** fits (≈1M-token context, tuned for fast lightweight summarization), whereas Whisper (audio), Ada-002 (embeddings), and Phi-3-Mini (small context) don't. A **serverless endpoint** is the low-operations, pay-per-token way to deploy it, and the **Foundry Chat playground** provides the interactive pre-integration testing the requirement calls for. All three community members give the identical triple.

More AI-300 practice questions