AI Runtime
24.3k 2026-04-24
mozilla-ai/llamafile
Distribute and run LLMs with a single file, no installation required.
Core Features
Single-file distribution and execution of LLMs
No installation, runs locally on most OS/CPU architectures
Includes whisperfile for single-file speech-to-text
Based on llama.cpp and Cosmopolitan Libc for broad compatibility
Supports recent LLM models and functionalities
Quick Start
curl -LO https://huggingface.co/mozilla-ai/llamafile_0.10.0/resolve/main/Qwen3.5-0.8B-Q8_0.llamafile && chmod +x Qwen3.5-0.8B-Q8_0.llamafile && ./Qwen3.5-0.8B-Q8_0.llamafileDetailed Introduction
llamafile is a Mozilla.ai project designed to significantly enhance the accessibility of open Large Language Models (LLMs) for both developers and end-users. It achieves this by packaging llama.cpp and Cosmopolitan Libc into a single, self-contained executable. This innovative approach eliminates complex installation procedures, allowing users to run LLMs locally on diverse operating systems and CPU architectures with unparalleled ease. Additionally, it integrates whisperfile for single-file speech-to-text capabilities, further broadening its utility.