Python backend for My-GPT.
- CMake and Make
- Python 3.8+ and Pip 3
- RabbitMQ
- CUDA Toolkit and related drivers (if your system has a GPU)
- Orca-Mini-3B model file
git clone https://github.com/aadarsh-ram/my-gpt-backend.git
cd my-gpt-backend
python -m venv venv
. venv/bin/activate
chmod +x ./requirements-setup.sh
./requirements-setup.sh
- Download the model file, convert it to GGUF format using this script.
- Update .env file with the path to the model file in the
MODEL_PATH
variable. Make any necessary changes for the RabbitMQ server. - Run the backend with
python3 app.py