llamafile
Distribute and run an LLM as a single executable file.
- Star 数
- 26.3K
- Fork 数
- 1.7K
- 未关闭 Issue
- 213
- 最近提交
- 2026-10-12
数据来自 GitHub API · 更新于 2026-10-15
llamafile 是什么？
llamafile from Mozilla combines llama.cpp with Cosmopolitan Libc so a model and its runtime ship as one file that runs on Windows, macOS, Linux and BSD — download, make executable, run.
主要功能
- One file runs on most operating systems
- Built-in web chat UI and OpenAI-compatible API
- Fast CPU inference
系统要求
- 平台
- Windows · macOS · Linux
- 内存
- Depends on the model
如何安装 llamafile
根据项目 README 整理。开源项目更新很快，请以 GitHub 页面上的最新说明为准。
macOS / Linux
- 1Download a .llamafile from the README’s model list, make it executable and run it.
chmod +x Llama-3.2-1B-Instruct.Q6_K.llamafile ./Llama-3.2-1B-Instruct.Q6_K.llamafile
Windows
- 1Rename the downloaded file to end in .exe and run it (Windows limits .exe files to 4 GB).
优势
- Easiest distribution format
- No dependencies
不足
- 4 GB executable limit on Windows
llamafile 替代品对比
| 软件 | Star 数 | 许可证 | 平台 | 显存 |
|---|---|---|---|---|
| llamafile Distribute and run an LLM as a single executable file. | 26.3K | Apache-2.0 | — | |
| Ollama Get up and running with large language models locally. | 183.2K | MIT | — | |
| llama.cpp LLM inference in C/C++ — the engine behind most local AI apps. | 131.2K | MIT | — | |
| KoboldCpp Single-file llama.cpp-based runner for GGUF models with a story-writing UI. | 12K | AGPL-3.0 | — | |
| Jan Open-source ChatGPT-style desktop app that runs 100% offline. | 45K | AGPL-3.0 | — | |
| GPT4All Private, on-device chat with local document Q&A. | 77.4K | MIT | — |
常见问题
llamafile 免费吗？
是的——llamafile 是开源软件，采用 Apache-2.0 许可证。商用前请确认软件及模型权重的许可条款。
llamafile 支持 Windows、Mac 和 Linux 吗？
llamafile 支持：Windows, macOS, Linux。
运行 llamafile 需要什么配置？
RAM: Depends on the model
如何安装 llamafile？
1. Download a .llamafile from the README’s model list, make it executable and run it.
llamafile 有哪些好的替代品？
常见替代品包括 Ollama, llama.cpp, KoboldCpp, Jan，详见本页对比表。