Gemini Nano

Google

Gemini Nano is Google's most efficient Gemini model, built to run directly on-device on Android through AICore and the ML Kit GenAI APIs. It targets latency-sensitive, privacy-focused, and offline tasks (summarization, smart reply, rewrite, proofreading) without a network round-trip, and pairs with cloud Gemini in a "hybrid inference" pattern that routes heavier reasoning to the server.