AI Stays Local

DocumentationTrust

Privacy model

Local inference happens on your machine. That is a statement about where the work runs, not a promise that the software never uses a network, and the difference matters.

What stays local

  • Inference: the model runs on your hardware.
  • Your documents, and the index built from them.
  • Your prompts and the answers to them.

What may use the network

  • Downloading a model in the first place.
  • Product updates.
  • Any integration you turn on deliberately.

We do not claim that nothing ever leaves your machine. In the documented local mode, with no cloud integration enabled, inference and document retrieval run locally. That is a checkable statement, unlike an absolute.

The notice covering this website and the beta form is separate: privacy notice.

Last updated 2026-09-22T10:05:46-04:00 · Request access

These pages describe the contract: ports, endpoints, volumes, the permission model and the behaviour you can expect. The command reference for the CLI is published with the build each tester receives, so that it matches the binary in their hands rather than a page written ahead of it.

Available through a controlled private beta. Larger model profiles are added as they complete hardware validation.