Windows

How to install the llama.cpp libraries on Windows.

Choose the tab for the processor that you want to use.

Decide where you want to put the library files, then run this command:

yzma install --lib C:\path\to\lib

To use an NVIDIA GPU, first install the CUDA drivers. See the CUDA installation guide.

Then run this command:

yzma install --lib C:\path\to\lib --processor cuda

The installer also downloads the cudart files that CUDA on Windows needs.

To use an Intel CPU, GPU, or NPU with OpenVINO, run this command.

yzma install --lib C:\path\to\lib --processor openvino

The download has the OpenVINO runtime. A GPU or an NPU also needs the Intel drivers. See Configurations for Intel hardware.

OpenVINO uses the CPU unless you choose another device:

set GGML_OPENVINO_DEVICE=GPU

To use an AMD GPU, run this command:

yzma install --lib C:\path\to\lib --processor rocm

To use Vulkan, first install the Vulkan SDK.

Then run this command:

yzma install --lib C:\path\to\lib --processor vulkan

The yzma install command prints instructions for your system when it finishes. Follow them to complete the installation.

Then set the YZMA_LIB environment variable:

set YZMA_LIB=C:\path\to\lib

You do not need to add this directory to PATH. yzma loads the DLLs that the libraries need from the same directory.

Next steps