ESP32-AI-Chatbot

Simple LLM Assistant Chatbot using Ollama API For ESP32. It offers a simple command-line interface to send questions and receive responses from the Ollama Large Language Model.

Features

• Dynamic Preprompt Customization: Users can modify the preprompt context to change the AI's behavior or persona, making the interactions as flexible as needed.
• Secure Configuration Management: Utilizes environment variables to manage API configuration securely, ensuring sensitive information is kept out of the source code.
• Robust Error Handling: Includes mechanisms to gracefully handle network issues and API errors, ensuring the application remains responsive under various conditions.

Requirements

• WiFi Library
• HTTPClient Library
• ArduinoJson Library
• Internet connection to access the Ollama LLM API.

Installation

1. Clone the Repository

   git clone https://github.com/Den-Sec/ESP32-AI-Chatbot/
   cd ESP32-AI-Chatbot/
   

2. Install Libraries Install the required libraries via the Arduino IDE:

   Navigate to Sketch > Include Library > Manage Libraries 
   

3. Set Environment Variables Update the following constants at the top of your code:

   OLLAMA_URL=http://your-ollama-url:port
   MODEL_NAME=your-model-name
   ssid=your-ssid
   password=your-password
   
   

Replace http://your-ollama-url:port and your-model-name with your actual Ollama API URL and model name.

Usage

• Upload the script to your ESP32.
• Open the Serial Monitor.
• Send messages by typing them into the monitor.

To end the session, type 'exit'.

Contributing

Contributions to this project are more than welcome!

License

Distributed under the MIT License. See LICENSE for more information.