Question 1
You have a Microsoft Foundry project that contains a high-traffic agent. After a recent update, operational costs increase significantly. Monitoring confirms that the volume of user traffic to the agent remains unchanged. The model deployment and its pricing are the same. You need to determine whether the agent is consuming more input tokens, output tokens, or both. Some model calls include long tool results in the prompt. Which observability capability should you use?
- A.latency
- B.evaluation metrics
- C.run success rate
- D.token usage