The path to near-perfect digital cloning. Fully open source.

About

PersonalGPT is a cutting-edge project that leverages state-of-the-art AI to create an immersive digital clone. It offers a unique and interactive experience across various mediums, harnessing advanced multimodal image recognition and text generation capabilities.

Core Features:

• GPT-4o Support: OpenAI's most advanced, multimodal flagship model that’s cheaper and faster than GPT-4 Turbo. It supports native vision with images, offering top-tier performance and efficiency.
• Configurable Memory: Features a robust aiocache queue, applying a First In, First Out (FIFO) strategy for consistent and lossless data management.
• Fully Asynchronous Processing: Efficiently handles high volumes of concurrent messages, maintaining seamless performance.
• Customizable Parameters: Provides comprehensive customization through the Discord Developer Portal and an intuitive configuration file.

Note

The active-dev branch is focused on active feature development.

🧪 Getting Started (Beta)

To get started with PersonalGPT, follow these steps:

1. Download the repository

   • Clone using Git:

   git clone https://github.com/praetor29/personalgpt.git

   • Or download the latest archive.

2. Modify configuration files
   Customize config.yaml.example and prompts.yaml.example with your settings. Remove the .example extension from the file names.

3. Add API tokens
   Add your OpenAI token and Discord App token to .env.example, then rename the file to .env.

4. Install Python
   Ensure you have Python 3.12.3 installed on your system.

Tip

Using a virtual environment like Miniconda makes managing dependencies easier and less messy.

5. Install required packages
   Run the following command to install the necessary Python packages:

   # navigate to the project directory
   cd personalgpt
   # install all requirements
   pip install -r requirements.txt

6. Run the application

   This runs PersonalGPT in the foreground.

   • Launch script:

   # navigate to project directory
   cd personalgpt
   # make launch script executable (first time)
   chmod +x ./personalgpt.sh
   # run launch script
   ./personalgpt.sh

   • Directly:

   python -m src

7. Run in the background (Linux)
   To run PersonalGPT in the background, use nohup:

   nohup <launch script or python command> &

Note

The bots in the dev server are deployed using docker compose. However, the Dockerfile is not yet available in the repository. Releasing a docker image may be explored in the future.

---

Contributions

⚡ Development

praetor29

✨ Beta Testing

yumeshu, jinzou, goldiereal, vamp1044, ambasinghhh

🔗 Useful Links

• Project Webpage (not updated to v3.0)
• Join our Discord

🌐 Core Technologies

PersonalGPT utilizes several open-source libraries and technologies:

• discord.py: A modern, easy to use, feature-rich, and async ready API wrapper for Discord written in Python.
• openai-python: Official OpenAI library for AI interaction. SDK v1 compliant.
• aiocache: Asyncio cache supporting multiple backends (memory, redis and memcached).

Note

Support for ElevenLabs Voice Cloning/Discord voice support were dropped in v3.0. Voice support may be re-implemented in a later release.

⚖️ License

This project is licensed under the GNU General Public License v3.0.