Tool Calling

Tool calling is a four-step process:

1. Available tools are declared in the context;
2. The model itself decides which tool to call and with what parameters;
3. The framework executes the tool and adds the result to the context;
4. The model decides what to do next.

The model is responsible for decision-making, the framework for execution.

This transforms an LLM from a text generator into a system capable of action.

Related: ReAct cycle, Parameter passing fidelity