Google AI Edge Gallery. Offline AI in your phone! – AI News – #2 June 2025

Imagine this situation: you’re traveling in a place with limited internet access, and you need to quickly translate a text, write a creative email, or analyze a photo. Until now, this meant one thing – a problem. Most powerful AI tools like ChatGPT, which we use daily, require a constant connection to the cloud and its immense computing power. However, Google has decided to break this barrier, giving us a glimpse of the future with its new experimental project – AI Edge Gallery.

The idea of running advanced AI models entirely on our phones sounded like science fiction until recently. It required enormous resources that smartphones simply didn’t have. Until now.

Google AI Edge Gallery allows you to download and run language models locally, meaning that after a one-time download, you can use them anytime, completely offline. This is a real revolution in the accessibility of this technology. No more frustrating connection error messages when you need AI support.

Think of AI Edge Gallery as a kind of laboratory where we can test various possibilities of local AI. The heart of the system is lightweight, open generative AI models from Google, such as Gemma. These have been optimized to work efficiently on the hardware resources of a typical smartphone. The app is currently available for Android, but will soon be available for iOS as well.

It should be noted that AI Edge Gallery is still an experimental project. You won’t find it in the official Google Play store. To install it, you need to download the APK file directly from the project page on GitHub. The process requires a bit of effort – you need to allow installation of apps from unknown sources in your phone’s settings. You’ll also need an account on the Hugging Face platform, which serves as a hub for AI models. Although it sounds a bit complicated, this is a process that even less advanced users will have no trouble with, and more importantly, it provides great satisfaction.

At the start, Google offers three main features:

Moving AI from the cloud to edge devices, also known as edge computing, means greater privacy (your data doesn’t leave the phone), lower latency, and unprecedented accessibility. Projects like Google AI Edge Gallery represent the democratization of AI in its purest form, which in the future could completely change the way personal assistants work. While today it’s still an experiment, the trend is clear – the future of AI is personal, accessible, and always at hand, even when you’re offline.

Do you want to stay up-to-date with the latest news from the world of artificial intelligence? Follow Delante’s news and subscribe to our newsletter!

Source of information: https://github.com/google-ai-edge/gallery