Harbor

Harbor is an open source tool that sets up a complete local LLM stack with a single command. It wraps a large catalog of containerized services behind one command line interface and Docker Compose, so that inference backends, chat front ends, web search, evaluation and other AI tooling come prewired to work together instead of being configured by hand. The repository includes integrations for services such as Ollama, SearXNG, Perplexica and Promptfoo, along with profiles, routines and an app. It is aimed at developers, homelab users and self hosters who want to run and compare local models and related services on their own hardware. The project is written largely as shell tooling around containers, is distributed through package managers and the GitHub repository, and has a community Discord server. Harbor is released under the Apache 2.0 license and is free to use. The listed URL is the project's GitHub repository, and the only costs involved are those of the hardware or hosting the user runs it on.