llama.cpp

llama.cpp

Docker 应用程序 from grtgbln's Repository

概述

Inference of Meta's LLaMA model (and others) in pure C/C++

要求


        The image for this container is several gigabytes. If you receive a "no space left on device" warning during installation, please increase the vDisk size in your Docker settings.
        This container expects a "model.gguf" file to be present in the model storage path.
        If you are using an Nvidia GPU, add "--gpus all" to the Extra Parameters field under Advanced.
    

运行时参数

网络用户界面
http://[IP]:[PORT:8000]/
网络
bridge
特权
false

模板配置

WebUIPorttcp

Container Port: 8000

目标
8000
默认值
8000
价值
8000
Model Storage PathPathrw

Storage for model

目标
/models
默认值
/mnt/user/appdata/llama_cpp/model
价值
/mnt/user/appdata/llama_cpp/model

详细信息

存储库
ghcr.io/ggml-org/llama.cpp:full
最后更新2026-05-31
初见2026-02-24

在Unraid 上运行 llama.cpp 。

llama.cpp 已被列入Unraid OS 的社区应用程序。探索Unraid ,构建灵活的家庭服务器、NAS 或家庭实验室。