ForgeLLM is built upon the exceptional work of the MLX and MLX-LM projects. We want to express our sincere gratitude to:
-
The MLX Team at Apple for creating the MLX framework, which provides the core machine learning capabilities that power ForgeLLM.
-
The MLX-LM Team for developing MLX-LM, which forms the foundation of our language model training and inference capabilities.
-
The MLX Community for their continuous contributions, support, and for building an ecosystem of models and tools that enhance the capabilities of MLX.
ForgeLLM does not reimplement the core functionalities of MLX or MLX-LM, but rather builds upon them to provide convenient features, parameter management, monitoring, model handling, and other utilities that make working with language models more accessible.
Our project would not be possible without the groundbreaking work of these teams and communities. We are committed to acknowledging their contributions and adhering to their licensing terms.
We also acknowledge the various open-source libraries and tools that ForgeLLM depends on, including but not limited to:
- Flask and Flask-SocketIO for web interface capabilities
- Various Python libraries that enable our data processing and visualization features
Please see our requirements in the README.md and pyproject.toml for a complete list of dependencies.