Microsoft Azure AI Engineer Associate AI-102 Practice Question

The Assistants API defines three main objects. An Assistant stores the system instructions and the list of tools it can call. A Thread is a long-lived container that stores every message exchanged and the results of any tool calls. A Run is a transient execution of the assistant against a specific thread. Because the requirement is to preserve conversation history across multiple requests, you must create a Thread once and reuse its identifier for each subsequent Run. Creating only a Run or a new Assistant for every turn would lose prior messages, and an Azure OpenAI deployment alone provides no session state.