# v_llama_cpp

A V language binding for llama.cpp, letting V programs run quantized large language models such as DeepSeek or Qwen locally on CPU hardware.

v_llama_cpp is a V-language binding for llama.cpp, the C++ LLM inference framework, letting V projects call llama.cpp functionality directly. It inherits llama.cpp's CPU-only inference, INT4/INT5/INT8 quantization support, and cross-platform reach across Windows, Linux, macOS, and mobile devices, so models such as DeepSeek, Qwen, or ChatGLM can run locally on consumer hardware.

Installation clones the repository (from GitHub, or mirrors on Atomgit and Gitee for users in China) and runs `v install.vsh`, which builds llama.cpp if it is not already present; a matching uninstall script is provided. Example programs in `./examples/` show basic usage, and a first run automatically downloads a default GGUF model file, with Hugging Face and ModelScope listed as model sources.

Website: https://github.com/sakana-ctf/v_llama_cpp