Awesome

upython-chat-gpt

ChatGPT for MicroPython.

Overview

This repository contains an example that allows you to connect to ChatGPT using MicroPython. It makes a POST request to the OpenAI API with a question and receives a response that is then printed on an OLED display.

This example has been tested to work with the Giga R1 WiFi board and a generic ESP32 board and should be compatible with other boards supporting MicroPython.

To connect to OpenAI, you will need an account with a payment plan as this is not a free service. However, you can play around with this for days without spending more than a dollar.

Hardware Requirements

• GIGA R1 WiFi
• OLED SSD1315 (I2C) display.*

*Other similar OLED displays may work but have not yet been tested with this setup.

Hardware Setup

Wire the display to the GIGA R1 WiFi using the following chart:

Software Setup

1. Install MicroPython on your board. To install on the GIGA R1 WiFi, follow this guide.
2. Install a MicroPython-compatible editor (like Thonny or Arduino Lab for MP)
3. Install the SSD1306 OLED driver on your board, either by installing directly via Thonny's built-in package manager or by copying over the ssd1306.py file in this repository to the lib folder of your board.
4. Create an account with OpenAI. Once created, you will need to register a credit card or another payment method. You should then be able to create an API key.
5. In the example script, add your API key in the api_key field.
6. In the example script, you will also need to add your WiFi network SSID + PASS.

You should now have met all the requirements to use the example script provided in this repository.

Usage

When you have set it up properly, anytime you launch the script, the following will happen:

• The content of open_ai_question is sent in a POST request to the OpenAI API, which will return a response. This uses the urequests & ujson modules.
• The response is printed on the OLED display, letter by letter for extra dramatic effect.*

*I couldn't find a text wrap function in the ssd1306.py module so I made a quick and dirty improvisation (the print_oled() function).