llama.cpp joins Hugging Face

GGML (the llama.cpp development team) has joined Hugging Face (HF) to drive the sustainable development of local AI. Georgi Gerganov and his team will take on the role of expanding and supporting the llama.cpp and ggml communities at HF.

Key areas of collaboration:

• Unifying model definition and inference: Strengthen the integration between HF's transformers library and llama.cpp, optimizing the workflow so that new models become immediately usable in llama.cpp as soon as they are defined.
• Improving user experience (UX): Improve the packaging of ggml-based software to build an environment where even general users can easily deploy and use local models.

The project will continue to operate as 100% open source and community-driven as before, with HF providing resources for the project's long-term growth.