Run AI on your own machine.

Chat with a model, ask questions about your own documents, search the web privately, make images, let it read web pages for you — all on your own computer. Nothing leaves it.

Download for Windows macOS early Intel Mac early Linux early Free · Nothing to install first · Windows 10/11, macOS (Apple Silicon or Intel), Linux
Your system will warn you about the download. The installers aren't signed yet, so Windows shows "Windows protected your PC" and macOS blocks the first open (System Settings → Privacy & Security → Open Anyway). That is about a certificate we haven't bought, not about the file. Checksums are on the release page.

What it does

Chat

A capable model running on your own machine. On first run it picks one that fits your hardware and tells you why; you can install others from the Models page.

Ask your documents

Drop in a PDF or a folder of notes and ask questions about them. They are indexed on your computer and never uploaded anywhere.

Private web search

Search the web from the chat without a search engine watching. Needs Docker Desktop, which is the one optional install.

Make and edit images

Generate pictures and edit the ones you have, locally. Optional, and the one part with a manual setup, since the model files are large.

Let it browse

It can open a real web page and read it to answer you. It only clicks or types when you switch that on, per conversation.

See the test results

Every tool ships with what it scored — on which model, on what date — and the tests are included, so you can rerun them on your own hardware instead of taking our word for it.

Before you install

The first run downloads a few gigabytes

The engine that runs models is 1.4 GB on Windows, and the model it chooses for you is between 1.4 and 13 GB depending on what your machine can hold — it names the size on screen before it starts. Measured end to end on a clean Windows 11 machine with no graphics card: installer to chat in 6m 04s, about 3 GB downloaded. Afterwards it starts in seconds.

It works without a graphics card, slowly

No card is required — it was tested on a machine with none. But the card is what makes it quick, and it decides how large a model you get: more video memory means a larger, more capable model; with none, it uses a small one on the processor. It tells you which it chose and why, and the Models page will install any of the others.

Windows is the tested one

The Windows installer has been run from scratch on a clean machine. The macOS and Linux downloads are the same application built for those systems; they exist and they are expected to work, and nobody has yet installed one on a clean machine and said so — that is what early means above. macOS comes in two builds — Apple Silicon (2020 and later) and Intel; Apple menu → About This Mac says which you have, and the wrong one refuses to open rather than running badly. Linux is an AppImage; a .deb is on the release page too. If one of them doesn't work for you, the release page is the place to say so.

What it needs

Graphics
None required; a card with 8 GB or more gets you a much larger model
Memory
8 GB
Disk
10 GB, considerably more for large models and image generation
System
Windows 10 or 11 · macOS, Apple Silicon or Intel · Linux (macOS and Linux are early)
Install first
Nothing. Docker Desktop is optional, and only if you want private web search