Harbor
FreeA containerized toolkit for running local LLM backends, UIs, and supporting services with one command. #opensource
FreeFree tier
About Harbor
Harbor is a CLI and companion app that lets you spin up a complete local LLM stack—backends like Ollama, llama.cpp, or vLLM, frontends like Open WebUI, plus supporting services like SearXNG for web search, Speaches for voice chat, and ComfyUI for image generation—all pre-wired to work together with a single harbor up command. No manual setup: just pick the services you want and Harbor handles the Docker Compose orchestration, configuration, and cross-service connectivity so you can focus on actually using your models.
Key Features
One-command deployment of a complete local LLM stack via `harbor up`
Pre-wired integration between backends (Ollama, llama.cpp, vLLM), frontends (Open WebUI), and supporting services (SearXNG, Speaches, ComfyUI)
Docker Compose orchestration with automatic configuration and cross-service connectivity
Supports agentic modules via `harbor launch --workflow` and Boost workflows
Includes `harbor doctor` for troubleshooting and `harbor how` for documentation
Active development with frequent releases and an integration test suite
Pros & Cons
Pros
- Single-command setup eliminates manual configuration
- Pre-wired services work together out of the box
- Wide choice of backends, frontends, and supporting services
- Open-source and free to use
- Containerized for isolation and easy cleanup
- Active community and regular updates
Cons
- Requires Docker experience for custom setups
- Relatively new tool with occasional integration issues (first-boot failures noted in v0.5.4)
- May be overkill for users who need only a single LLM frontend or backend
- Documentation and service catalog still evolving
Best For
Running a local LLM stack for development and experimentationBuilding applications with web RAG, text-to-speech, and speech-to-textVoice chat and image generation with local modelsAgentic workflows and Boost agent modules for complex tasksExploring and testing different LLM backends and frontends without manual configuration
FAQ
What is Harbor?
Harbor is a CLI and companion app that lets you spin up a complete local LLM stack—backends like Ollama, llama.cpp, or vLLM, frontends like Open WebUI, plus supporting services such as SearXNG for web search, Speaches for voice chat, and ComfyUI for image generation—all pre-wired to work together with a single `harbor up` command.
How do I install Harbor?
Installation guides are available in the repository, including an agent install prompt for Claude Code, Codex, and similar tools. The CLI and companion app can be installed via the provided shell or PowerShell scripts.
What services are supported by Harbor?
Harbor supports a growing catalog of services including backends (Ollama, llama.cpp, vLLM, MLX, oMLX, Daytona), frontends (Open WebUI), and supporting services (SearXNG, Speaches, ComfyUI, DMR, and more). The full catalog is maintained in the Harbor Services documentation.