February 2024

Built from the ground up for heterogeneous graphs where types and relations are represented by distinct sets of nodes and edges.

Build graph neural networks with TensorFlow GNN 1.0

Analyze complicated relationships between objects using deep learning. Build with TensorFlow GNN 1.0 , a production-tested library that provides tools to process relational data at scale and train models easily.

Read the blog

Solve OGBN-MAG with TensorFlow GNN

Try an interactive notebook to solve Open Graph Benchmark's Node classification task. Learn how to obtain manageable-sized inputs and train models with minimal coding.

Run Colab

Try Gemma, a new family of open LLMs

Explore lightweight, state-of-the art open models built from the same technology used to create Gemini models. You can run Gemma with a TensorFlow backend through Keras 3.

Build with Gemma

Automate web AI model testing in a Colab

Learn how to create an automated testing environment for your browser-based, client-side models by using NVIDIA GPUs in Colab, headless Chrome, and Puppeteer.

Read the doc

Run Colab

Apply to lead a Google Developer Student Club

Leading a GDSC is a great opportunity to learn programming skills, dive deep into Google technologies and create local impact, while also building your network. 2024 applications are open now for India and start March 1, 2024 for other regions.

Learn about GDSC

Stay Connected