Directory Structure

This document provides an overview of how the llama.cpp Windows documentation is organized.

content/llama-cpp-windows/
├── index.md                       # Landing page & Overview
├── DIRECTORY_STRUCTURE.md         # Map of the docs
├── installation/                  # Setup & Backend-specifics
│   ├── prerequisites.md           # VS Code, CMake, Git, Build Tools, etc.
│   ├── cpu-installation.md        # Standard build
│   ├── cuda-installation.md       # NVIDIA GPU setup
│   └── amd-installation.md        # AMD GPU (HIP) setup
├── inference/                     # Running models
│   ├── model-formats.md           # GGUF, downloading models (Hugging Face)
│   ├── cli-usage.md               # Command line arguments & examples
│   └── prompt-templates.md        # Chat templates (Alpaca, Llama-3, etc.)
├── configuration/                 # Environment & Settings
│   ├── environment-variables.md   # Path setup & system vars
│   └── optimization.md            # Quantization & performance tips
└── references/                    # External links
    └── external-resources.md      # Official repo, CUDA docs, etc.

How to Navigate

• Installation: Start here if you are setting up your environment for the first time.
• Inference: Go here once you have a built llama.cpp and want to start chatting with models.
• Configuration: Use these guides if you need to fine-tune your performance or environment.
• References: Use these for official documentation and additional learning.

---

Last Updated: 2026-05-03