llamafile
A Mozilla.ai project that packages a large language model and its runtime into a single executable file, so it runs locally on most operating systems and CPU architectures with no installation.
llamafile lets a user distribute and run large language models as a single file. It combines llama.cpp with Cosmopolitan Libc into one framework that collapses the complexity of running an LLM down to a single executable, called a llamafile, that runs locally on most operating systems and CPU architectures with no installation.
It also includes whisperfile, a single-file speech-to-text and translation tool built on whisper.cpp using the same packaging approach. Versions from 0.10.0 onward use a new build system that tracks recent llama.cpp releases and models more closely, while older releases remain available for anyone who prefers the previous behavior.
It is a Mozilla Builders project, now maintained by Mozilla.ai, written in C++. On Windows, only executables under 4GB can run, and the downloaded file needs an .exe extension added before running.