What does an agent have that a plain model deployment does not?
Why D is the answer
Instructions and tools that let it take actions
An agent combines a model with instructions and tools such as file search.
In detail
- Instructions and tools let an agent take actions.
- An agent can still make mistakes.
Use case
An illustrative scenario, not a real incident.
A chat deployment answers questions. An agent can also take actions, such as searching files, because it has instructions and tools.
See it in different ways
flowchart TD A["Chat completion"] --> B["One prompt, one response"] C["Agent"] --> D["Instructions and a model"] D --> E["Tools and knowledge"] E --> F["Steps toward a goal"]
sequenceDiagram participant U as User participant A as Agent participant T as Tool U->>A: Request A->>T: Uses a tool, such as file search T-->>A: Result A-->>U: Answer or completed action
mindmap
root(("Agent"))
n1["Instructions"]
n2["What it should do"]
n3["Tools"]
n4["What it can use"]Key point: Instructions and tools let agents act.
Why the other options are wrong
- A. A guarantee that it will never make a mistake
- No AI system can guarantee this.
- B. A faster processor on the host machine
- An agent does not imply faster hardware.
- C. A shorter list of supported languages
- This is not what defines an agent.