Private AI, light enough to fly.

Three states. Radically lighter.

Ordinary AI does its sums in fine-grained numbers — expensive to store, expensive to move, expensive in energy. ternii's models think in just three: +1 0 −1. That makes the whole assistant small enough, and frugal enough, to run offline on a phone — no data centre, no bill, no waiting on a network.

In our app, on an iPhone 17 Pro, the difference is not subtle.

One phone, one app build, one prompt, only the model changed. Ours is ternary from training and runs on the GPU; a 4-bit model is a compressed copy of a bigger one and has no GPU path in our app — so this is an as-it-ships comparison, and we say so.

Don't take our word for it. Watch the phone.

Live screen recordings. Every number traces to a bench.

Nothing lost to compression.

Most phone AI is a big model squeezed down to fit — and it loses accuracy in the squeeze. ternii's model was trained in three states, so the model on your phone is the model we trained. There is no compression step, and nothing to lose in one.

A small model, given the right facts at the right time.

A tiny model doesn't know everything — so we don't ask it to. ternii's trick is augmentation over size: routing and retrieval put curated facts and live tools in front of the model exactly when a question needs them, and the model writes from those facts only.

Your phone is the whole computer.

ternii is private by construction, not by promise. There's no account to create, no profile to build, and no server that sees your conversations, your notes, or your files. In airplane mode it works exactly the same.

Experts, downloaded only if you want them.

Knowledge lives in small, curated expert packs — one per subject. The app stays light and you add the packs you care about; each is 55–334 KB and installs in a tap. New packs appear over time, with no app update needed.

Routines that do the legwork.

A routine gathers from the tools you've connected — calendar, weather, news, prices, your notes — and hands the model only the real data to summarise. No invented meetings, no made-up headlines: if it isn't there, it isn't in the briefing. One tap, a useful answer.

What ternii can reach.

Every tool is listed with what it actually does and where the data goes. Green means nothing leaves the phone. Only the "open web" tools touch the internet, and each sends the minimum — a city name, a feed address, a ticker. Anything that acts (an alarm, a message, a call) is handed to the phone's own app for you to confirm. This is the Android list; iPhone matches except where noted.

Your name, in the weights.

Put your name — or a message in your words — on the Founding Roll. If you opt in, it is trained into the next Brain: part of the model, in every copy, on every phone that runs it. The roll's fingerprint goes into every model file we ship, and your entry is issued a certificate on a public blockchain that names the model back. £10 buys 500 million training tokens.

As far as we can find, the first AI model file that names its own founders and its certificate contract — with each certificate naming the model back.

Names on the roll, as they land.

Every approved entry, in roll order, shown the way its member chose. This is the roll whose fingerprint goes into the model file — the same list, on every phone that runs it.

A clubhouse, not a mailing list.

Founding Members get a sign-in link by email — no password. Behind it: the work, as it happens, and a board where you tell us what to build.