Models
Pull open-weight models once and run them offline, or switch to a hosted model when a task is worth it.
Models, offline and private
0:48
How the 1,000+ catalogue is filtered, and which models run on your own machine.
Local or cloud
A local model runs entirely on your hardware. There is no network round trip, no per-token cost, and nothing about the request leaves the machine. A cloud model is faster and stronger on hard problems, and costs tokens.
You can mix them freely in one session. The model picker marks each entry LOCAL or CLOUD so you always know which you are about to use.
Getting models
•
lk models — list everything available, local and hosted.
•
lk pull <model> — download an open-weight model for offline use.
•
lk use <model> — set the default for this and future sessions.
•
/models — the same picker from inside a chat session.
A pulled model is yours. It keeps working with the network unplugged, and it does not expire.
Choosing one
•
Small local models (3B–8B) are fast, fit comfortably in memory, and are enough for edits, explanations and tests.
•
Larger local models (14B–70B) are noticeably better at multi-step reasoning if you have the RAM and patience.
•
Hosted models are the right call for genuinely hard problems, long context, or when you need the strongest answer available.
If you are unsure, start with the default. Memory is the constraint that actually decides what runs well: a 3B model is comfortable on a laptop, an 8B wants real headroom, and lk pull --list shows the download size of each before you commit to it.
Running fully offline
Once a model is pulled and you have signed in at least once, no network is required. Conversations are kept locally and sync when you reconnect; nothing is lost in the meantime.
Each device also holds a signed offline token allowance for hosted models. When you reconnect, whatever you did not use goes back to your balance.
All documentation
Installation
Install the CLI on macOS, Linux or Windows, and sign in.
Setup, start to finish
The whole path on one page: install, sign in, pick local or hosted, get an answer.
Connecting accounts
One account everywhere, registering devices for the SMS bridge, and OAuth integrations.
Using models on every client
What runs locally in a terminal, a browser tab, a phone and the desktop app.
Commands
Every terminal command and every slash command, each with its own page.
Editor integrations
VS Code, Cursor, JetBrains, Neovim, Sublime, Zed, Godot, Unity, gh and git hooks.
The SMS bridge
Send a task to your own machine from your phone, and get the answer back.
The apps
One React Native codebase behind iPhone, Android, web and desktop — and how to run it.
Automation Studio
Idea to production: flows, per-step models, your machines or ours, live editing.
Enterprise seats
Licensed team seats under one owner account — invites, billing, and how seats resolve.
API reference
API keys, authentication, chat completions, pricing and errors — for your own software.
Notifications
Four channels, all opt-in, how to change them, and the limits on re-engagement.
Privacy
What leaves your machine, what never does, and how anonymisation works.
Troubleshooting
When a model will not load, a command hangs, or the CLI cannot be found.