Harbor

What is Harbor?

Harbor is a command-line tool and companion app for running a whole local LLM stack without wiring each piece by hand. A command such as harbor up starts a chosen set of services through Docker Compose and connects them: inference backends like Ollama, llama.cpp and vLLM, frontends like Open WebUI, and extras such as ComfyUI for generating images, SearXNG for web search and Speaches for voice chat. On macOS it can also run Docker Model Runner, MLX or oMLX on the host with Metal acceleration. The harbor launch command points installed coding tools, including Codex, Claude, OpenCode and VS Code, at a local backend, and Harbor Boost lets you script research and audit workflows around a model. Other commands cover config profiles, a QR code to open services on your phone, a built-in tunnel, and an eject command that exports a standalone docker-compose file. The repository lists no price, and the project is open source on GitHub with a Discord community.

Use cases

• Run local LLMs with a single command
• Connect a local backend to Open WebUI
• Add web search and RAG to local models
• Point coding agents at a local model
• Chat with a local model by voice
• Generate images locally with ComfyUI
• Script multi-step LLM workflows

Who it's for

• Developers
• Self-hosting enthusiasts
• Machine learning hobbyists
• AI engineers

Why use Harbor

• One command replaces hand-written Docker Compose files and per-service configuration.
• Services come already connected, so search, voice and image tools work inside the chat UI.
• Your models and data stay on your own hardware.
• The eject command exports your current setup, so you can leave Harbor without losing it.
• Config profiles make it easy to switch between model settings for different tasks.

Harbor features

• One-command harbor up service orchestration
• Backends including Ollama, llama.cpp and vLLM
• Frontends such as Open WebUI and LibreChat
• SearXNG web search wired to many services
• Speaches voice chat with speech-to-text and text-to-speech
• ComfyUI and Flux image generation
• harbor launch for coding tools like Codex and OpenCode
• Harbor Boost workflow scripting
• Config profiles and command history
• QR code and tunnel access to services
• Eject to a standalone Docker Compose file
• MCP management with MetaMCP

Harbor integrations

• Ollama
• llama.cpp
• vLLM
• Open WebUI
• SearXNG
• ComfyUI
• LibreChat
• Dify
• n8n
• LangFlow
• Flowise
• Perplexica

How to get started with Harbor

1. Install the Harbor CLI or companion app by following the installation guide.
2. Run harbor up with the services you want, such as a backend and Open WebUI.
3. Harbor generates the Docker Compose configuration and connects the services to each other.
4. Open the chat UI, or use harbor launch to send a coding tool to your local backend.
5. Save settings as a profile, or eject to a standalone Compose file when you want your own setup.

Harbor limitations

• Requires Docker for the container-based services
• The built-in tunnel exposes your machine to the internet and needs authentication set up
• Running many backends at once is not practical on typical hardware
• No hosted version or support plan is described

Harbor pricing

Harbor is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

Harbor alternatives

The main Harbor alternatives listed on Debuly are Ollama, LM Studio, Open WebUI, LocalAI, Jan and AnythingLLM. The table compares pricing and platforms; order is A–Z.

1. 

   AnythingLLM

   Open-source AI app for chatting with your documents and running agents, on desktop, in Docker or as a hosted team workspace.

   Chat assistantsAutomation and agentsDeveloper platforms
   Open sourceVisit AnythingLLM (official site)
2. 

   Jan

   Free, open-source desktop AI assistant that runs models offline on your computer or connects to cloud models with your own keys.

   Chat assistantsDeveloper platforms
   Open sourceVisit Jan (official site)
3. 

   LM Studio

   Desktop app for running open language models on your own computer, now with the Bionic agent, a local API and optional US-hosted cloud models.

   Chat assistantsDeveloper platformsCodingProductivity
   FreemiumVisit LM Studio (official site)
4. 

   LocalAI

   MIT-licensed open-source runtime that serves local models through OpenAI- and Anthropic-compatible APIs for text, images, audio and agents.

   Developer platformsAutomation and agentsChat assistants
   Open sourceVisit LocalAI (official site)
5. 

   Ollama

   Open-source app and CLI for running open AI models locally on macOS, Windows and Linux, with optional paid cloud models.

   Developer platformsCodingChat assistants
   Open sourceVisit Ollama (official site)
6. 

   Open WebUI

   Self-hosted AI chat platform for Ollama and OpenAI-compatible models with RAG, agents, tools, notes, channels and role-based access.

   Chat assistantsDeveloper platformsAutomation and agents
   Open sourceVisit Open WebUI (official site)

Harbor FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: none collected yet, so a monogram is shown.
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.