Tool Use / Function Calling
Definition
Function Calling allows a model to suggest tool calls that an application executes in a controlled manner.
A model answer identifies a tool and its arguments. The connected application checks permissions, executes the call, and returns the result. The model itself does not automatically have access to an API or database.
Simply explained
Like an assistant who not only talks, but can use a tool in a targeted manner.
Why it matters in practice
Helps to realistically classify the possibilities, quality and limitations of AI systems.
Example from business
Possible example: A support assistant reads an order status from a system after approval.
Limits and pitfalls
- Allow extensive permissions or unchecked tool results.
Origin & history
Current model APIs describe function calling as a multi-step process: tool proposal, execution in the application and return of the result to the model.
Use in mid-sized companies
- Possible application in medium-sized businesses: A support assistant reads an order status from a system after approval.
Frequently asked questions
What does tool use and function calls mean?
Function Calling allows a model to suggest tool calls that an application executes in a controlled manner.
What errors should be avoided?
Allow extensive permissions or unchecked tool results.
Sources
- Function calling: Modellvorschlag und Ausführung durch die Anwendung · OpenAI, Function calling: Modellvorschlag und Ausführung durch die Anwendung
Sources checked on 09.10.2026. Editorial application examples are hypothetical; no evidence of real customer results.
Editorially reviewed on: