Everything, one screen.

Say it or type it, and it carries out multi-step tasks across your Windows apps — long jobs that run while you do something else.

Windows · voice or text

Speech becomes text on your own device. Your audio never leaves it, and once it is set up it works offline.

Windows, Linux, Android

Voice to text on our servers: nothing heavy on your machine, no word cap, and audio and transcripts are not kept after delivery.

Windows, Linux, Android

Live speech translation in your ear, on your own Google Gemini key. Audio goes straight from your phone to the AI provider on your key — no account, and nothing of ours in between.

Android · Google Gemini

Indexes a repository into a code graph and hands AI agents exactly the context they ask for, instead of re-reading files.

Self-hosted · MCP + HTTP

Runs 4-bit language models on an ordinary CPU — pure Rust, no GPU, no Python.

CPU-only · Windows, Linux, macOS

Your agent works with placeholder keys while a deny-by-default proxy injects the real ones, only for hosts you allow, and logs every request it handles.

Windows-first

Opens the documents you already have — free, no account, no ads.

Android