Answers · straight, sourced, applies to anyone
Answers about running AI privately, on your own machine
The real questions people ask about private, local, and sovereign AI — answered directly, with every claim linked to a page that proves it or a primary source. Written and reviewed by people, not generated. Where the honest answer is a limit or a "not yet", it says so; that is what makes it worth citing over the marketing.
- What is the best small local model to run on my own machine?2026-09-01
- How do I know what data an AI app sends to the cloud?2026-09-01
- Does Ollama store my conversations and memory?2026-09-01
- What is the difference between local, private, and sovereign AI?2026-09-01
- How much RAM do I need to run a local LLM?2026-09-01
- Can I run a 120B model on a gaming GPU?2026-09-01
- How do I move my ChatGPT or Claude history to a local AI?2026-09-01
- How do I give a local LLM long-term memory?2026-09-01
- Is running AI locally actually private?2026-09-01
House rules. Every answer links or cites its claims. Where a claim is about our own product, the mechanism behind it is named and the limits are stated — the same rule as the Sovereignty Ledger. These pages exist to be quoted, with or against us; the test we hold ourselves to is at /three-questions.