Attach a PDF, TXT, Markdown, HTML or CSV file and ask about it. Retrieval and embeddings run on-device, and every answer links back to the passage it came from, so you can check the source.
A local model’s knowledge stops at its training cutoff, which is fine until you ask about this week. Private Mind pulls results from the web and reasons over them on your device, at the moments you decide it should.
Dictate a message and Whisper transcribes it on the phone itself. The model does the work instead of a transcription service, so what you say goes straight into the chat and nowhere else.
Save system prompts as reusable presets and give each model its own default. One preset can be a blunt code reviewer that skips the preamble, another a patient tutor that explains fractions to your kid.
