CLI + companion App to spin up complete local LLM stacks
Harbor saves you dozens of hours configuring your local AI stack. One command spins up LLM backends, chat frontends, web search, voice, image gen, fine-tuning, agent tools - all pre-configured to talk to each other. Manage your downloaded models, create config profiles, use agentic coding tools with your local LLMs, and much more.
Been working on this for almost two years. Harbor started because I was tired of spending evenings debugging Docker Compose files every time I wanted to try a new local LLM project. It was supposed to be a weekend script, but the problem turned out to be bigger than I thought and it grew into a catalog of 129 services.
The core idea is that local AI services should work together out of the box. harbor up gives you Open WebUI and Ollama. You add more services by name, like harbor up searxng speaches comfyui, and Harbor handles the wiring between them. SearXNG is already connected to your chat UI. Voice already works. Image gen already works. No Compose files to write, no ports to remember.
The catalog covers a lot of ground. There are 16 chat and creative frontends (Open WebUI, LibreChat, SillyTavern, ComfyUI and others), 21 LLM backends (Ollama, llama.cpp, vLLM, SGLang, KTransformers and others), and 90+ satellites for web search, voice, RAG, vector databases, workflow engines like n8n and Dify, fine-tuning with Unsloth, eval harnesses, monitoring, and more.
Everything runs locally, talks to each other, and is one harbor up away. The latest release adds harbor launch, which is the feature I'm most excited about. It runs coding agents like Claude Code, Codex, Copilot, and OpenCode against your local stack without touching a single env var. harbor launch --web gives them web search too and spins up the whole pipeline behind the scenes. Harbor also exposes an OpenAI-compatible API, so any tool or SDK code that speaks OpenAI can point at your local models as a drop-in replacement.
MIT licensed, no telemetry, everything stays on your machine. CLI and desktop app for Linux and macOS. If you run local models and are tired of gluing things together manually, give it a try.
Happy to answer questions about the architecture, supported services, or how specific workflows work.
About Harbor on Product Hunt
“CLI + companion App to spin up complete local LLM stacks”
Harbor launched on Product Hunt on May 27th, 2026 and earned 70 upvotes and 1 comments, placing #30 on the daily leaderboard. Harbor saves you dozens of hours configuring your local AI stack. One command spins up LLM backends, chat frontends, web search, voice, image gen, fine-tuning, agent tools - all pre-configured to talk to each other. Manage your downloaded models, create config profiles, use agentic coding tools with your local LLMs, and much more.
On the analytics side, Harbor competes within Open Source, Artificial Intelligence and GitHub — topics that collectively have 579.4k followers on Product Hunt. The dashboard above tracks how Harbor performed against the three products that launched closest to it on the same day.
Who hunted Harbor?
Harbor was hunted by Ivan Charapanau. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
For a complete overview of Harbor including community comment highlights and product details, visit the product overview.
Ivan Charapanau
Hey PH 👋
I'm Ivan, Harbor's author.
Been working on this for almost two years. Harbor started because I was tired of spending evenings debugging Docker Compose files every time I wanted to try a new local LLM project. It was supposed to be a weekend script, but the problem turned out to be bigger than I thought and it grew into a catalog of 129 services.
The core idea is that local AI services should work together out of the box. harbor up gives you Open WebUI and Ollama. You add more services by name, like harbor up searxng speaches comfyui, and Harbor handles the wiring between them. SearXNG is already connected to your chat UI. Voice already works. Image gen already works. No Compose files to write, no ports to remember.
The catalog covers a lot of ground. There are 16 chat and creative frontends (Open WebUI, LibreChat, SillyTavern, ComfyUI and others), 21 LLM backends (Ollama, llama.cpp, vLLM, SGLang, KTransformers and others), and 90+ satellites for web search, voice, RAG, vector databases, workflow engines like n8n and Dify, fine-tuning with Unsloth, eval harnesses, monitoring, and more.
Everything runs locally, talks to each other, and is one harbor up away. The latest release adds harbor launch, which is the feature I'm most excited about. It runs coding agents like Claude Code, Codex, Copilot, and OpenCode against your local stack without touching a single env var. harbor launch --web gives them web search too and spins up the whole pipeline behind the scenes. Harbor also exposes an OpenAI-compatible API, so any tool or SDK code that speaks OpenAI can point at your local models as a drop-in replacement.
MIT licensed, no telemetry, everything stays on your machine. CLI and desktop app for Linux and macOS. If you run local models and are tired of gluing things together manually, give it a try.
Happy to answer questions about the architecture, supported services, or how specific workflows work.