Models you pick, hosted or on your computer

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

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 model share --model qwen3 --name qwen --auth # https://qwen.USER.DOMAIN/v1, password-protected

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 credential and nothing to pay (On Your Computer).
A provider's model, with your own credential. Anthropic, OpenAI or OpenRouter, and your agent never holds the credential (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 model builders. A model you're building gets its own address, for your agent and for a demo (For Model Builders).

Pick where it runs

lnk model status, usage and request actions need the model namespace of a running agent. Outside that context, Link gives a usage message with the next commands, without offering to report a bug. Use lnk agent model status for your agent or lnk model list for runtimes on this computer.

Select your agent's model

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

	Costs	Who sees your words
On your computer	Nothing	No one, it never leaves this computer
A model provider	What the provider bills you	The provider
Shared	Nothing	Your computer, and whoever you give the password
For model builders	Nothing	Your 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.

~/.lnk/models.toml holds the addresses you add, and nothing else.
~/.lnk/model-facts.toml holds the models you describe with lnk model show.
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.

More about it

Every flag is in lnk model <command> --help. Security has what's protected and the gaps, and Decisions why it works this way. When it finds no model: Troubleshooting.

Program API

The model actions API gives agents and sandboxed programs the same request and answer shapes.

Docs

Configuration: The files lnk model keeps, and what each one holds
On Your Computer: A model on your computer is free, and nothing you say leaves it
For Model Builders: Give the model you're building its own address
Interface: What other programs can rely on from lnk model
Model actions: An agent's model has one API, whatever serves it
Context Reporting: See what your agent's context is spent on, part by part, for every harness: its instructions, each tool, and the conversation down to each tool's calls and results
A Model's Details: A program in a sandbox asks a model's details in one shape, whatever runs the model
Model Provider: Use Anthropic, OpenAI or OpenRouter with your own credential
Share a Model: Share the model on your computer from anywhere: your phone, your box or a friend
Security: What's protected: the user's model runtime (Ollama, LM Studio, ...) when it's served or exposed, and their machine from what discovery finds
Decisions: Why the models plugin works the way it does
Troubleshooting: What to do when lnk model finds no model, or finds more than one (Models)