Apeirogon runs open-source models entirely on your device. No cloud, no API calls, no third party server. The model runs locally, which means it works as fast as your hardware allows and never sends anything outside your machine.

Apeirogon is designed to run on modern consumer hardware. A Mac with apple silicon or a windows machine with a decnet GPU will give you the best experience. We'll be transparent about minimum specs before you download.

Never. Apeirogon processes everything locally. Your files, your prompts, your results stay on your machine. We have no pipeline receiving your data because we choose not to build one.

Cloud tools like ChatGPT and Claude send your data to remote servers on very request. Apeirogon runs entirely on your device, which means no usage caps, no monthly bills, no surveillance, and no internet connection required.

Yes. Once installed, Apeirogon works completely offline. On a plane, at a remote site, anywhere without a signal. Your work doesn't wait for Wi-Fi.

No subscriptions, no per-query costs, no usage tiers. We'll share pricing details closer to launch. What we can say now is that we didn't want to build a product that works against you.

Apeirogon itself is a commercial product, but it runs on open-source model weights. You're not locked into a proprietary model controlled by a third party. The foundation is transparent by design.