MLX Chat is an AI chatbot interface using MLX and FastHTML.
- MLX for LLM inference on Apple silicon
- FastHTML user interface
- Multiple AI model options
- Adjustable temperature and max tokens
- Conversation history
- Dark and light theme
-
Clone the repository:
git clone https://github.com/yourusername/mlx-chat.git cd mlx-chat
-
Install the package and its dependencies:
pip install -e .
To run MLX Chat, use the following command from the terminal:
mlx_chat
- Select an AI model
- Set temperature and max tokens
- Type your message
- Send and view the AI's response
MLX Chat uses MLX for on device LLM inference and FastHTML for the user interface.
- Clear Messages
- Stop Generation
- Light/Dark Mode
- Model Search