{
  "slug": "llama-cpp",
  "name": "llama.cpp",
  "category": "Runtimes / Inference",
  "tagline": "The gold-standard C/C++ inference engine for GGUF models, with Metal as a first-class backend.",
  "kit_says": "This is the engine under a lot of other engines. Version 0.6.0 got its own episode on the channel: new Metal code and a 'will it fit?' estimator.",
  "homepage": "https://llama.app",
  "repo": "https://github.com/ggml-org/llama.cpp",
  "docs": "https://github.com/ggml-org/llama.cpp/blob/master/docs/build.md",
  "license": "MIT",
  "platforms": [
    "macos-apple-silicon",
    "linux",
    "windows"
  ],
  "mac_studio_fit": "Primary",
  "what_it_is": "LLM inference in C/C++. Runs GGUF models from Hugging Face, includes a CLI and an OpenAI-compatible server with a built-in web UI. README: 'Apple silicon is a first-class citizen - optimized via ARM NEON, Accelerate and Metal frameworks'.",
  "gap_role": "Maximum control over quantization, context and sampling; widest GGUF model coverage.",
  "prerequisites": [
    "macOS/Linux shell or Windows PowerShell"
  ],
  "install": {
    "macos-apple-silicon": [
      {
        "run": "curl -LsSf https://llama.app/install.sh | sh",
        "note": "official installer from the README"
      },
      {
        "run": "# alternative: brew install llama.cpp   (Homebrew formula is at 0.6.0 as of 2026-10-15)"
      }
    ],
    "linux": [
      {
        "run": "curl -LsSf https://llama.app/install.sh | sh"
      }
    ],
    "windows": [
      {
        "run": "irm https://llama.app/install.ps1 | iex",
        "note": "PowerShell"
      }
    ]
  },
  "configure": [
    {
      "run": "llama cli -hf ggml-org/Qwen3.5-0.8B-GGUF",
      "note": "download from Hugging Face and chat (README example)"
    },
    {
      "run": "llama serve -hf ggml-org/Qwen3.5-0.8B-GGUF",
      "note": "OpenAI-compatible API server + web UI"
    }
  ],
  "verify": [
    {
      "run": "command -v llama",
      "expect": "path to the llama binary (Homebrew/older builds may expose llama-cli / llama-server instead)"
    },
    {
      "run": "curl -s http://localhost:8080/v1/models",
      "expect": "JSON listing the loaded model (server running; 8080 is the server's usual default, check startup log)"
    }
  ],
  "rollback": [
    {
      "run": "brew uninstall llama.cpp",
      "note": "if installed via Homebrew"
    },
    {
      "run": "# script install: follow uninstall instructions at https://llama.app"
    },
    {
      "run": "# -hf downloads are cached (macOS default ~/Library/Caches/llama.cpp); delete to free space"
    }
  ],
  "caveats": [
    "CLI naming changed across versions (llama cli / llama serve vs llama-cli / llama-server); use whichever your install provides."
  ],
  "works_with": [
    "open-webui",
    "litellm",
    "aider"
  ],
  "sources": [
    "https://github.com/ggml-org/llama.cpp",
    "https://formulae.brew.sh/formula/llama.cpp",
    "https://www.youtube.com/watch?v=pWu3cNd0Qyo"
  ],
  "sponsored": false,
  "verified_on": "2026-10-15",
  "gap_score": {
    "value": null,
    "status": "pending test on Mac Studio",
    "baseline": "frontier reference (see /about#gap-score)"
  },
  "benchmarks": {
    "status": "pending test on Mac Studio",
    "rig": "Mac Studio, Apple Silicon, 128GB unified memory",
    "results": []
  },
  "hardware_tested": [],
  "repo_meta": {
    "stars": 131267,
    "stars_as_of": "2026-10-15",
    "last_push": "2026-10-15",
    "archived": false,
    "license_spdx": "MIT"
  },
  "stack_layer": "runtime",
  "summary": {
    "what": "The engine behind most local AI.",
    "why": "Full control over speed, context and quantization."
  }
}