Base
Hardened build
minimal ports, security patches applied at build time
llama.cpp Server on Ubuntu 24.04
Azure
Applications
llama.cpp Server, an efficient local inference engine for open LLMs, ready to serve models from first boot.
Overview
llama.cpp is a high efficiency inference engine that runs open large language models on CPU and GPU hardware, exposing an OpenAI compatible server. It is used to host and serve local LLMs with minimal overhead.
Why the cloudimg image
cloudimg ships the llama.cpp server hardened and fully patched on Ubuntu 24.04 LTS, preconfigured so the inference endpoint is ready on first boot. 24/7 support is included with every image.
Common uses
- Local LLM inference
- Private model serving
- OpenAI compatible endpoints
See it running
Real screenshots taken while testing this image against its deployment guide.
More in Applications
Actual Budget Self-Hosted Sync Server
Applications
This image runs the Actual Budget 26.6 sync server, the open source local-first personal finance and envelope budgeting ...
Apache Answer
Applications
This image runs Apache Answer, the open source question and answer platform for teams at any scale.
Audiobookshelf Audiobook and Podcast Server
Applications
This image delivers Audiobookshelf 2.35, the popular open source self-hosted audiobook and podcast media server, fully i...
Baikal CalDAV & CardDAV Server
Applications
Baikal is an open source self-hosted CalDAV and CardDAV server built on the proven Sabre/dav library.
BookStack Knowledge Base - Ready-to-Run
Applications
BookStack is a free, open source platform for organising and storing documentation and knowledge.
Calibre-Web Self-Hosted Ebook Server
Applications
This image runs Calibre-Web from a dedicated Python virtual environment, served on the loopback interface and fronted by...