HarborDeploy your entire AI infrastructure with a single command

Harbor

Harbor is a powerful CLI and companion application that lets you deploy a complete local LLM stack with a single command. Spin up backends like Ollama, llama.cpp, or vLLM, frontends like Open WebUI, and supporting services including SearXNG for web search, Speaches for voice chat, and ComfyUI for image generation—all pre-configured to work together seamlessly.

Product Highlights

• One-Command Deployment: Launch your entire LLM infrastructure with harbor up—no manual Docker configuration needed
• Pre-Wired Integration: All services automatically connect to each other, eliminating complex setup procedures
• Multi-Backend Support: Choose from 15+ inference engines including Ollama, vLLM, llama.cpp, TabbyAPI, and SGLang
• Rich Ecosystem: Access 80+ services spanning UIs, backends, satellites, and workflow tools
• Coding Agent Integration: Use harbor launch to wire Harbor backends into Claude, Codex, and other coding agents

Use Cases

• Local AI Development: Build and test LLM applications without cloud dependencies or API costs
• Private AI Workflows: Keep all data on-premises with fully self-hosted RAG, chat, and image generation
• Coding with Local Models: Power coding agents like Claude Code and Codex with local OpenAI-compatible backends
• Voice-Enabled AI: Deploy speech-to-text and text-to-speech capabilities for conversational interfaces
• Research & Experimentation: Quickly switch between models and configurations for benchmarking

Target Audience

Harbor is designed for developers, AI researchers, and privacy-conscious users who want to run large language models locally without wrestling with Docker Compose files and manual service configuration. Ideal for those comfortable with command-line tools seeking a streamlined local AI infrastructure.