Google I/O

Gemma models: Unveiling the latest advancements

Gemma is Google’s family of lightweight, state-of-the art open models built from the same research and technology used to create the Gemini models. Join this session to learn how to get started. We’ll cover the types of models available and how to customize them to excel at new tasks. You’ll learn how to perform training and inference with Keras 3.0 in just a few lines of code, how to deploy models to Android with MediaPipe, and how to serve them on Google Cloud.

Technical session
Join a community group

Meet developers, discover local groups, and build your global network.

Get started
I/O Connect

Explore, network, and get hands-on with the latest products.