mozilla-ai/llamafile

Distribute and run LLMs with a single file.

What it solves

llamafile simplifies the distribution and execution of Large Language Models (LLMs) and speech-to-text tools. It removes the complexity of installation, dependency management, and environment setup, allowing users to run open LLMs locally on most operating systems and CPU architectures without needing to install software.

How it works

It combines llama.cpp and whisper.cpp with Cosmopolitan Libc to create a single-file executable. This "llamafile" bundles the model weights and the inference engine together into one portable, cross-platform binary that can be run directly on macOS, Linux, BSD, and Windows (with some size limits on Windows).

Who it’s for

Developers and end users who want to run open LLMs or speech-to-text tools locally with zero installation and minimal technical friction.

Highlights

  • Single-file distribution: Bundles model and engine into one executable.
  • Cross-platform: Runs on most operating systems and CPU architectures.
  • Zero installation: No need to install dependencies or manage environments.
  • Includes whisperfile: Provides a single-file tool for audio transcription and translation based on whisper.cpp.
  • Broad model support: Supports various model sizes and architectures through its integration with llama.cpp.

Related

  • Project
    ggml-org/Llama-macOS

    A macOS menu bar app for running local LLMs using llama.cpp, featuring zero-configuration setup and hardware-aware model recommendations.

  • Project
    SciSharp/LLamaSharp

    A cross-platform C# library based on llama.cpp that allows developers to run LLaMA and other LLMs locally on CPU and GPU with high-level APIs.

  • Project
    ngxson/wllama

    A WebAssembly binding for llama.cpp that enables running LLMs, multimodal models, and tool calling directly in the web browser using WebGPU and Wasm.

  • Project
    nobodywho-ooo/nobodywho

    An on-device inference engine that enables developers to run LLMs, speech-to-text, and text-to-speech locally and offline across multiple platforms.

  • Project
    withcatai/node-llama-cpp

    A Node.js library that provides pre-built bindings for llama.cpp, enabling developers to run LLMs locally with automatic hardware acceleration and TypeScript support.