AetherCompute

Lend your graphics card to AI

Your computer's graphics card runs AI language models (the kind behind chat assistants) for other people while it would otherwise sit idle.

This is a test for now. Your computer connects, sets itself up and shows up on our side. No real requests are sent to it yet, and nothing is paid.

One small file, under 0.1 MB. The app finds our server by itself, so there is nothing to type in.

Then

1. Unzip it. On Windows, right-click it and choose "Extract All"; starting it from inside the zip won't work. A Mac usually unzips it for you.
2. Start it:
   • Windows: double-click start.bat. If a blue "Windows protected your PC" box appears, click "More info", then "Run anyway". If Windows asks whether to run the file, click "Run".
   • Mac: double-click start.command. The first time, macOS stops it because it was downloaded: open System Settings, then Privacy & Security, scroll down, click "Open Anyway" next to start.command, then "Open". After that, double-clicking just works.
   • Linux: run ./start.sh in the folder.
3. Leave the window open. The first time, it sets itself up: its own copy of Python, then llama.cpp, the program that runs the models, in the build that suits your graphics (about 12 MB on a Mac, 30 MB for AMD or Intel graphics, 580 MB for an NVIDIA card). That takes a few minutes. After that it starts in seconds.

What you need

• A graphics card with 8 GB of its own memory, or a Mac with an Apple chip (M1 or newer) and 16 GB of memory. Other computers can join the test too; they just won't be given models to run.
• About 2 GB of free disk. You don't need to install anything, not even Python.

What it does to your computer

• Everything it downloads is checked against a fingerprint before it is used.
• Nothing is installed system-wide, nothing appears in your applications, and any Python you already have is left alone.
• To remove it, delete the folder, then the hidden .aether folder in your home folder, which holds llama.cpp and any models it downloaded.