Private. Offline. Yours.

LocalInference runs powerful open AI models entirely on your phone. No account, no cloud, no tracking — your conversations never leave your device.

On-device. Chat works fully offline after a one-time model download.

Zero data collection. No analytics, no crash reporting, no servers. We couldn't read your chats if we wanted to.

A real assistant. Tools, web search, contacts & calendar (always asks first), vision, read-aloud.

For tinkerers. 16 curated models from 0.5B–8B, live tok/s stats, bring-your-own API keys.

Coming soon to the App Store & Google Play