Skip to content
AImpact
IT EN
High Foundation Models · 1 min read

Gemini Nano on-device: frontier LLM directly on the phone

In one sentence Google DeepMind deploys Gemini Nano (1.8B and 3.25B) on Pixel 8 Pro and Galaxy S25, offline execution on NPU via Android AICore API. First time a frontier lab puts an LLM directly on the device.

Verified Official source
ShareLinkedInX
Reading level

Gemini Nano is the smallest version of Google's Gemini family, designed to run directly on smartphones without needing an internet connection.

It runs on the NPU (Neural Processing Unit), a specialized chip present in modern phones like Pixel 8 Pro and Galaxy S25, which allows AI computations to be done very efficiently without draining the battery.

For the first time, a major AI lab has put its own model directly inside a phone as a built-in feature, not as a separate app. This means private, fast AI that works even offline.

Companies

Google DeepMind

Tools

Gemini Nano, Android AICore

Tags

Gemini NanoGoogle DeepMindOn-Device AINPUAndroid

Sources