Back to Cookbook
BeginnerEdge / Local 9 min read

llama.cpp on Windows with CUDA

Build llama.cpp with NVIDIA GPU support on Windows 11 — the path of least resistance for PC gamers.

Windowsllama.cppCUDAGGUF

Prerequisites

Install Visual Studio Build Tools, CMake, and CUDA Toolkit 12.x.

powershell
winget install Kitware.CMake
# CUDA: download from developer.nvidia.com/cuda-downloads

Build

Enable CUDA backend during CMake configure.

powershell
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp
cmake -B build -DLLAMA_CUDA=ON
cmake --build build --config Release -j
Deployment guides are educational. Each model is subject to its own license — read the official Hugging Face model card before downloading or deploying.