Local-first AI

Your datacenter.The one you already own.

Own the model. Own the memory.The AI runs on your machine. After installation, nothing leaves it.

Free during the beta — no account, no sign-up.

120B on a laptop with an 8 GB GPU (measured in June 2026, internally, with a version before 1.0) 0 bytes leave your machine after installation (on first launch, memown downloads its engine from GitHub; the model is a file you download, for example from Hugging Face) Windows verified on September 23, 2026 (Windows 11) Linux Ubuntu 24.04 or newer (glibc ≥ 2.38), install libgomp1 macOS July 2026 archives, signed and notarized, not re-verified since
LOCAL
120B on a laptop with an 8 GB GPU (June 2026, internal test, before 1.0). Same answer after a restart. A local API for your apps.
In plain words

What memown does.

Big models.

In an internal test in June 2026, a version of memown before 1.0 ran a model bigger than the graphics card's memory, by also using the computer's RAM and disk.

Nothing leaves.

After installation, nothing leaves your machine: the AI runs locally. On first launch, memown downloads its engine from GitHub; the model is a file you download yourself, for example from Hugging Face.

It remembers.

Your context is saved in a file on your machine (memown.mem) and reused after memown restarts.

Proof

What we measured, and when.

120B

A model of about 120 billion parameters started on a laptop with an 8 GB graphics card and 16 GB of RAM, with a version of memown before 1.0, and was ready to answer. Its engine (llama.cpp), tested alone on the same laptop, answered slowly. Measured in June 2026, internally.

measured
byte-identical

After memown restarts, it reuses its saved context, and the answer is identical, byte for byte, to one computed from scratch on the same machine. Measured on September 23, 2026 (Windows 11 and Ubuntu 24.04).

measured
0 bytes

Leave your machine after installation. On first launch, memown downloads its engine from GitHub; the model is a file you download yourself, for example from Hugging Face.

by design
The studio

Engineering what comes next.

memown is built by ZMLabs, a deep-engineering studio in Sète, in the South of France.

Sète · Occitanie · Francecontact.zmlabs@proton.me
Ready when you are

Try it on your own computer.

On September 23, 2026, memown ran a small model on a Windows 11 virtual machine with 8 GB of RAM and no graphics card, and gave the same answer after a restart.