On Windows 11 workstations, WinGet keeps command-line tools in the same package-management flow as browsers, editors, and runtimes. The llama.cpp package gives local AI users prebuilt llama-cli.exe and llama-server.exe binaries without cloning the source tree or compiling CMake targets.

The community-maintained Windows Package Manager repository identifies the package as ggml.llamacpp and tracks upstream llama.cpp release archives. Installing by this exact identifier avoids substring matches with similarly named packages.

The package supplies Vulkan builds for both x64 and ARM64 Windows, with a matching Microsoft Visual C++ Redistributable dependency. WinGet selects the architecture automatically and makes the portable executables available on PATH.

Steps to install llama.cpp with WinGet:

  1. Open Windows Terminal with a non-administrator Windows PowerShell tab.

    A working WinGet installation from App Installer is required. A missing winget command means App Installer needs installation or repair before continuing.

  2. Refresh the winget package source.
    PS> winget source update --name winget
    Updating source: winget...
    ##### snipped #####
    Done
  3. Confirm the exact llama.cpp package identifier.
    PS> winget search --id ggml.llamacpp --exact --source winget
    Name      Id            Version
    -------------------------------
    llama.cpp ggml.llamacpp b11946

    The version changes as llama.cpp publishes new releases. The package identifier remains ggml.llamacpp.

  4. Install llama.cpp from the winget community repository.

    The installation adds executables to PATH and may install a Microsoft Visual C++ Redistributable dependency with administrator approval.

    PS> winget install --id ggml.llamacpp --exact --source winget

    On x64 Windows, the selected archive and runtime dependency use x64 instead of ARM64.

  5. Select Yes if User Account Control shows Microsoft Corporation as the Visual C++ installer's verified publisher.

    The dependency prompt appears when runtime installation requires administrator approval. After the dependency finishes, WinGet continues with the llama.cpp archive.

    ##### snipped #####
    Successfully installed
  6. Exit the installation PowerShell session.
    PS> exit
  7. Open a new Windows Terminal window from Start.

    The fresh window reads the updated PATH so the packaged commands can resolve.

  8. Confirm that WinGet records the installed package.
    PS> winget list --id ggml.llamacpp --exact --source winget
    Name      Id            Version
    -------------------------------
    llama.cpp ggml.llamacpp b11946
  9. Verify that llama-cli.exe starts.
    PS> llama-cli.exe --version
    version: 0.7.0-dev (build 11946, commit 8ca4060b7)
    built with Clang 20.1.8 for Windows arm64
  10. Verify that llama-server.exe starts.
    PS> llama-server.exe --version
    version: 0.7.0-dev (build 11946, commit 8ca4060b7)
    built with Clang 20.1.8 for Windows arm64

    The version, build, compiler, and architecture reflect the installed release and workstation. Model weights are separate from the binaries; inference and API requests require a GGUF model.
    Related: Run a local GGUF model with llama.cpp
    Related: Start the llama.cpp server

Share this guide