Local Server Using LLM API (compatible with OpenAI's functions)

A Python script that utilizes LM Studio as a local server that implements a tool-driven workflow with an LLM:

• The local function generate_receipt(order_id) produces a random list of items, each with a randomly assigned price.
• The LLM is tasked with summing the item prices itself to compute the final total.
• Tool registration: Exposes generate_receipt to the LLM via the OpenAI-compatible API.
• Model responsibility: The model receives only item names and prices; it must calculate and report the total.