Distribute and run LLMs with a single file.
About this project
llamafile is an open-source inference and training project on GitHub: Distribute and run LLMs with a single file. It is infrastructure rather than a finished product: the layer that serves, quantises, trains or orchestrates models. It is maintained by mozilla-ai and published as a public repository you can read, fork and run yourself.
On GitHub the repository is tagged cross-platform, gguf, llama-cpp, local-ai, local-inference, local-llm, open-source-ai, single-file-executable and speech-to-text, which is a fair summary of what it covers. That generally means an API surface, batching and memory management, and often an OpenAI-compatible endpoint so existing clients keep working unchanged.
It is written mainly in C++ and released under no standard SPDX license identifier — the repository uses a custom or modified license, so read it before commercial use. It is actively maintained: the most recent commit landed on November 9, 2026.
To try it, clone the repository and follow its README; that file is the canonical install guide. There is also a project page at https://docs.mozilla.ai/llamafile for documentation or a hosted demo. Because it runs on your own infrastructure, budget time for environment setup: dependencies, model weights and configuration usually take longer than the first successful run.
Key facts as recorded on GitHub: 26,666 stars; 1,684 forks; primary language C++; last commit on November 9, 2026; topics cross-platform, gguf, llama-cpp, local-ai, local-inference, local-llm.