> ## Documentation Index
> Fetch the complete documentation index at: https://amd-gaia.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Setup

> Install prerequisites for GAIA C++ agent development

<Note>
  **Looking for Python?** This page covers C++ prerequisites. For Python prerequisites (uv, Node.js), see the [Python Setup](/docs/setup).
</Note>

## System Requirements

| Requirement | Supported                                            |
| ----------- | ---------------------------------------------------- |
| **OS**      | Windows 11, Ubuntu 24.04+                            |
| **CPU**     | Any x86\_64 processor                                |
| **RAM**     | 8GB minimum, 16GB+ recommended                       |
| **Disk**    | \~500MB for build artifacts and fetched dependencies |

***

## Install Prerequisites

<Tabs>
  <Tab title="Windows">
    ### Step 1: Install a C++ Compiler

    You need a compiler with C++17 support. **Visual Studio 2019 or later** is recommended.

    Install the free **Visual Studio Community** edition with the "Desktop development with C++" workload:

    1. Download from [visualstudio.microsoft.com](https://visualstudio.microsoft.com/downloads/)
    2. In the installer, select **Desktop development with C++**
    3. Click Install

    Verify the compiler is available (open **Developer Command Prompt**):

    ```bat theme={null}
    cl
    ```

    <Tip>
      **Already have Visual Studio?** Make sure the "Desktop development with C++" workload is installed via the Visual Studio Installer.
    </Tip>

    ### Step 2: Install CMake

    CMake 3.14 or later is required.

    ```powershell theme={null}
    winget install Kitware.CMake
    ```

    Verify (restart your terminal if needed):

    ```powershell theme={null}
    cmake --version
    ```

    <Tip>
      CMake is also bundled with Visual Studio. If you selected the C++ workload, you may already have it — check with `cmake --version` first.
    </Tip>

    ### Step 3: Install Git

    Git is required by CMake FetchContent to download dependencies automatically.

    ```powershell theme={null}
    winget install Git.Git
    ```

    Verify (restart your terminal if needed):

    ```powershell theme={null}
    git --version
    ```

    ### Step 4: Install Lemonade Server

    The agent needs an OpenAI-compatible LLM server. [Lemonade Server](https://lemonade-server.ai) is recommended (optimized for AMD hardware).

    Download and run the installer (v10.10.0):

    ```powershell theme={null}
    # Download the MSI installer
    curl -L -o lemonade-server-minimal.msi https://github.com/lemonade-sdk/lemonade/releases/download/v10.10.0/lemonade-server-minimal.msi

    # Run the installer
    msiexec /i lemonade-server-minimal.msi
    ```

    Or download directly from the [Lemonade v10.10.0 release page](https://github.com/lemonade-sdk/lemonade/releases/tag/v10.10.0).

    After installation, restart your terminal and start the server:

    ```powershell theme={null}
    lemonade-server serve
    ```

    <Note>
      **Not using Lemonade?** Any OpenAI-compatible server works — llama.cpp, Ollama, vLLM, or a cloud endpoint. See the [Integration Guide](/docs/cpp/integration#using-alternative-llm-backends) for configuration examples.
    </Note>

    ### Optional: Install uvx (for MCP demos)

    The System Health Agent demo requires `uvx` to launch the Windows MCP server. If you only plan to use the Wi-Fi Troubleshooter or build your own agent, you can skip this.

    ```powershell theme={null}
    pip install uv
    ```

    Verify:

    ```powershell theme={null}
    uvx --version
    ```

    <Success>
      You're ready to build! Continue to the [Quickstart](/docs/cpp/quickstart).
    </Success>

    <Tip>
      **Having issues?** Check the [Troubleshooting](/docs/reference/troubleshooting) guide, [create an issue](https://github.com/amd/gaia/issues) on GitHub, or contact us at [gaia@amd.com](mailto:gaia@amd.com).
    </Tip>
  </Tab>

  <Tab title="Linux">
    ### Step 1: Install a C++ Compiler

    You need a compiler with C++17 support. **GCC 9 or later** is recommended.

    ```bash theme={null}
    sudo apt update
    sudo apt install build-essential
    ```

    Verify:

    ```bash theme={null}
    g++ --version
    ```

    ### Step 2: Install CMake

    CMake 3.14 or later is required.

    ```bash theme={null}
    sudo apt install cmake
    ```

    Verify:

    ```bash theme={null}
    cmake --version
    ```

    <Tip>
      If your distribution ships an older CMake, install a newer version from [cmake.org/download](https://cmake.org/download/) or via `pip install cmake`.
    </Tip>

    ### Step 3: Install Git

    Git is required by CMake FetchContent to download dependencies automatically.

    ```bash theme={null}
    sudo apt install git
    ```

    Verify:

    ```bash theme={null}
    git --version
    ```

    ### Step 4: Install Lemonade Server

    The agent needs an OpenAI-compatible LLM server. [Lemonade Server](https://lemonade-server.ai) is recommended (optimized for AMD hardware).

    Install via the Launchpad PPA (Ubuntu 24.04+):

    ```bash theme={null}
    sudo add-apt-repository ppa:lemonade-team/stable
    sudo apt-get update
    sudo apt-get install -y lemonade-server
    ```

    Or browse all platform options on the [Lemonade v10.10.0 release page](https://github.com/lemonade-sdk/lemonade/releases/tag/v10.10.0).

    After installation, start the server:

    ```bash theme={null}
    lemonade-server serve
    ```

    <Note>
      **Not using Lemonade?** Any OpenAI-compatible server works — llama.cpp, Ollama, vLLM, or a cloud endpoint. See the [Integration Guide](/docs/cpp/integration#using-alternative-llm-backends) for configuration examples.
    </Note>

    ### Optional: Install uvx (for MCP demos)

    The System Health Agent demo requires `uvx` to launch the Windows MCP server. If you only plan to use the Wi-Fi Troubleshooter or build your own agent, you can skip this.

    ```bash theme={null}
    pip install uv
    ```

    Verify:

    ```bash theme={null}
    uvx --version
    ```

    <Success>
      You're ready to build! Continue to the [Quickstart](/docs/cpp/quickstart).
    </Success>

    <Tip>
      **Having issues?** Check the [Troubleshooting](/docs/reference/troubleshooting) guide, [create an issue](https://github.com/amd/gaia/issues) on GitHub, or contact us at [gaia@amd.com](mailto:gaia@amd.com).
    </Tip>
  </Tab>
</Tabs>

***

## Prerequisites Summary

| Tool                                          | Minimum Version                      | How to Verify                             |
| --------------------------------------------- | ------------------------------------ | ----------------------------------------- |
| C++ Compiler                                  | C++17 support (MSVC 2019+ or GCC 9+) | `cl` (Windows) or `g++ --version` (Linux) |
| CMake                                         | 3.14+                                | `cmake --version`                         |
| Git                                           | any                                  | `git --version`                           |
| [Lemonade Server](https://lemonade-server.ai) | 10.10.0                              | `lemonade-server --version`               |
| uvx                                           | any (optional)                       | `uvx --version`                           |

<Note>
  All C++ library dependencies (nlohmann/json, cpp-httplib, Google Test) are fetched automatically by CMake during the build — no manual installs required.
</Note>

***

<small style="color: #666;">
  **License**

  Copyright(C) 2025-2026 Advanced Micro Devices, Inc. All rights reserved.

  SPDX-License-Identifier: MIT
</small>