#llm

Function calling (also called tool use) is a capability that lets an LLM output structured commands — like get_weather(location='Cairo') — which your own code then executes, bridging the gap between what the model says and what it can do.