OpenAI API Usage Guide

This tutorial introduces how to install various model services on the Module LLM and LLM630 Compute Kit, and how to access and invoke their model capabilities via the standard OpenAI API on PC or other devices. It covers common scenarios including chat completion, speech recognition, speech synthesis, etc., helping you build efficient AI edge applications.

1. Preparation

Before use, please refer to the respective device’s software upgrade tutorial to add the M5Stack apt repository info and update the index.

• Module LLM Image & Software Upgrade Tutorial
• LLM630 Compute Kit Image & Software Upgrade Tutorial

2. Dependency Updates

1. Access the device terminal via UART or SSH and run the following command to update the apt repository index.

apt update

2. Update the llm core package on the device.

apt install lib-llm

3. Update the related dependency packages.

apt install llm-sys llm-llm llm-vlm llm-whisper llm-melotts llm-openai-api

4. After installing the above packages, reboot the device.

reboot

3. Python OpenAI

On the PC, configure your Python environment and install the official OpenAI Python library via pip. Subsequent steps will use this library and the standard OpenAI API interface to access the LLM device and invoke its model capabilities.

pip install openai

4. Feature Index

• Model List
• Chat Completion
• Speech-to-Text
• Text-to-Speech
• Chatbox