Tool calling

Tool calling lets a model request that your application runs a function and returns the results back into the conversation.

How it works

1) You send a message describing available tools
2) The model chooses a tool and arguments
3) Your app executes the tool
4) You send the tool result back

Notes

Exact payload format depends on the model. Psyllama supports OpenAI-compatible flows, so many tool-calling clients can work with minimal changes.