A private assistant
that runs on your own machine.

Ulu runs open-weight language models directly on your computer. No API keys, no accounts to chat, and nothing you type leaves the machine. That holds on a plane, in a tunnel, or with the Wi-Fi switched off.

Download for macOS See how it works Apple silicon · macOS 13.3 or later · Free

Intel Macs are not supported. The iOS and Android apps are in development.

The model runs on your hardware

The weights sit in your machine's memory and generate on your own GPU.

Real acceleration

Metal on Apple silicon, OpenCL on recent Adreno, and an NPU path on newer Snapdragon. Everything else falls back to a genuinely usable CPU path.

Works with no signal

Inference is a local process talking over loopback. There is no request to fail, so the assistant behaves identically offline.

Yours once downloaded

Weights are ordinary files on your disk. Pick a model to match your machine, swap it whenever, and delete it when you want the space back.

It asks before it touches the network

Most apps claim privacy. Ulu makes it checkable.

There are only a few reasons Ulu ever opens a connection: fetching a model, signing in, checking a subscription. Each one asks first, names the exact host, and says what is being sent and what is not.

Every request is written to a log you can read in Settings. Your conversations are never among them.

Pick a model that fits

Bigger models reason better and run slower. Ulu tells you which ones your machine can actually hold before you spend the download, measured against the memory you have rather than a number printed on a page.

DesktopDiskMemoryDownloads
Qwen 3.5 4B3.2 GB8 GB5.4M
Gemma 4 E2B 2B active3.8 GB8 GB4.1M
Gemma 4 E4B 4B active5.6 GB16 GB4.9M
Qwen 3.5 9B6.1 GB16 GB7.3M
Gemma 4 12B6.8 GB16 GB2.6M
Qwen 3.8 27B16.2 GB32 GB7.6M
Gemma 4 26B A4B 4B active16.9 GB32 GB10.4M
Gemma 4 31B18.2 GB32 GB4.6M
Qwen 3.6 35B A3B 3B active21.5 GB32 GB7.1M
PhoneDiskMemoryDownloads
Qwen 3.5 0.8B508 MB2 GB1.9M
Qwen 3.5 2B1.2 GB3 GB1.3M
Qwen 3.5 4B2.6 GB8 GB5.4M
Gemma 4 E2B 2B active2.9 GB8 GB4.1M