A CLI tool to manage and execute LLaMA files.
- Initialize and download models
- Pull specific models by name
- Quickstart to execute a default model
- Serve models on the internet using ngrok
- Start models with specific configurations
- Python 3.7 or higher
- Pip (Python package installer)
- ngrok account (optional, for serving models)
Install the package using pip:
pip install solo-cli
solo-cli init
solo-cli pull model_name
solo-cli quickstart
solo-cli serve --port 8080
solo-cli start model_name --port 8080
solo-cli init
solo-cli pull llava-v1.5-7b-q4
solo-cli quickstart
solo-cli serve --port 8080
solo-cli start llava-v1.5-7b-q4 --port 8080
Typer Requests tqdm ngrok
Add support for more models Improve error handling Add more configuration options
Contributions are welcome! Please open an issue or submit a pull request on GitHub.