Level 3 — Tool Calling: The LLM Decides When and How to Act
~12 min read
A human defines a set of tools the LLM can access; the LLM now decides not just which one to use, but WHEN to use it and what arguments to pass — real, if still bounded, agency over action.
Level 3 — Tool Calling: The LLM Decides When and How to Act is a Pro topic
Sign in, then upgrade to Pro or Power to unlock this topic and the full AI Engineering curriculum.
Key points
- •A human defines the available tools and their interfaces; the LLM decides WHEN to use them and WHAT arguments to pass
- •This is a real step up from routing — it involves reasoning about necessity and constructing arguments, not just classifying into a fixed lane
- •Many requests correctly need zero tool calls — the LLM's judgment about whether to act at all is part of this level's capability
- •The human's role shifts from defining full decision logic to defining the toolset and interfaces the LLM operates within
- •This is what most people informally mean by 'the LLM can use tools' — the foundation the multi-agent and autonomous levels build on