問題文
A developer wants to try several pretrained chat models and compare their answers before writing any code. Which capability of the service is intended for this?
選択肢
- The Files API, which is where prompts are uploaded for evaluation and where the service stores the comparison report that it produces for each model in the tenancy.
- The Vulnerability Scanning service, which produces the side-by-side answer comparison once the candidate models are registered with it.
- A dedicated AI cluster, which must be created first because every comparison of two models requires reserved capacity for each of the models involved.
- The playground in the Console, which lets you test hosted pretrained and custom models and refine prompts and parameters without writing code.