Teaching AI to see the world more like we do

D4RT: Teaching AI to see the world in four dimensions

D4RT: Unified, efficient 4D reconstruction and tracking up to 300x faster than prior methods.

LoRA training scripts of the world, unite!

TTS Arena: Benchmarking Text-to-Speech Models in the Wild

How we’re bringing AI image verification to the Gemini app

We’re expanding our presence in Singapore to advance AI in the Asia-Pacific region

Google DeepMind opens a new Singapore research lab, accelerating AI progress in the Asia-Pacific region.

Harness, Scaffold, and the AI Agent Terms Worth Getting Right