Any model, on your computer or with your key

Run a model on your computer for free, or use Anthropic, OpenAI or OpenRouter with your own key. You can also share your computer's model, or one you're building, from anywhere.

  • Free and private, on your computer. lnk installs a model for you with Ollama, or finds the one you already run, and there's no API key and nothing to pay (On Your Computer).
  • A provider's model, with your own key. Anthropic, OpenAI or OpenRouter, and your agent never holds the key (Model Provider).
  • Shared from anywhere. Your computer's model reaches your phone, your box or a friend through a tunnel, behind a password (Share a Model).
  • For ML engineers. A model you're building gets its own address, for your agent and for a demo (For ML Engineers).

Quickstart

lnk up models tunnels                                  # the models plugin, and the tunnel to share them
lnk model install                                      # Ollama and the model that suits this machine (asks first)
lnk model list                                         # the models running here
lnk tunnel open qwen3 --name qwen --auth friend:secret # https://qwen.<you>.local.link/v1, password-protected

Pick where it runs

Switch your agent between them in one command: Switch Models.

CostsWho sees your words
On your computerNothingNo one, it never leaves this computer
A model providerWhat the provider bills youThe provider
SharedNothingYour computer, and whoever you give the password
For ML engineersNothingYour server, and whoever you give the password

What it needs

You need a Mac or a Linux computer. lnk up models adds the plugin, and sharing a model also needs the tunnel plugin. lnk model install picks a model on an Apple Silicon Mac, and elsewhere you name one. A model on your own hardware is free.

Where things live

Link keeps two small files, and your models stay in their runtime.

  • ~/.config/lnk/models.toml holds the addresses you add, and nothing else.
  • ~/.config/lnk/model-facts.toml holds the models you describe with lnk model facts.
  • Models stay in their runtime, in Ollama's own folder for the ones install pulls.
  • Ollama, when install adds it, is the official one. It's in /Applications/Ollama.app, or ~/Applications, on macOS, and a system service on Linux.
  • lnk plugin remove models removes the plugin, not Ollama or its models.

Troubleshooting

SymptomFix
lnk model list finds nothinglnk model install, or start your runtime on its default port.
A runtime on another portlnk model add <url>, or --upstream <url> for one command.
"several model runtimes are running"Name a model, or pass --upstream.
"lnk model is part of the models plugin, which isn't installed"lnk up models.

Every flag is in lnk model <command> --help. Security has what's protected and the gaps, and Decisions why it works this way.