Function Calling
Also known as: Tool use, Tool calling
The model emits structured JSON specifying a tool call instead of free text; the host executes it.
An LLM capability where the model outputs structured JSON specifying which tool/API to call with what arguments, rather than free-form text. The host application then executes the call and feeds the result back into the conversation.
In practice
Foundational to every agent system. Expect questions on schema design, parallel tool calls, and error recovery.
How it compares
Function calling is the model behavior; MCP is one standard protocol for how tools are described and invoked.
Comparisons that include Function Calling
Related topics
Questions that mention this term
- Flashcard: what does it mean when an LLM 'makes a tool call'?
- Your colleague says MCP and function calling do the same thing. Where are they wrong?
- Which statements correctly describe the MCP primitive types?
- Roles in a chat training example: pick the three that are standard.
- Tool calling fine-tune: data shape and the common pitfall
- In MCP, what exactly is a 'tool' and who decides when to call it?
Related terms
Agent Loop
The runtime: LLM call → tool call → observation → LLM call → … until a final answer or step cap.
AI Agents
LLMs that loop: plan → call tools → observe results → repeat until done.
Chain-of-Thought (CoT)
Ask the model to think step by step before answering. It boosts accuracy on reasoning tasks.
Context Rot
Long-running chats degrade: early instructions get forgotten, and tool calls become less reliable.
In-Context Learning (ICL)
Show the model a few examples in the prompt and it learns the pattern, no fine-tuning needed.
Model Context Protocol (MCP)
An open protocol (from Anthropic) for letting LLMs call tools and read resources from any compatible server.