MIT Technology Review I was able to try out Astra last week in a live demo behind closed doors. It was a breathtaking experience, but there’s a gap between polished promo and live demo.

Astra leverages Gemini 2.0’s built-in agent framework to answer questions and perform tasks via text, voice, image and video, and invoke existing Google apps such as Search, Maps and Lens on demand. “It brings together some of the most powerful information retrieval systems of our time,” says Bibo Xu, product manager at Astra.

Gemini 2.0 and Astra are joined by Mariner, a new agent built on top of Gemini that can surf the Internet for you; Jules, a new Gemini-powered coding assistant; and Gemini for Games, an experimental assistant that lets you chat and ask for tips while playing video games.

(And let’s not forget that last week Google DeepMind also announced Veo, a new video generation model; Imagen 3, a new version of its image generation model; and Willow, a new type of chip for quantum computers. Phew. Meanwhile, CEO Demis Hassabis was in Sweden yesterday and received his Nobel Prize.)

Google DeepMind claims that Gemini 2.0 is twice as fast as the previous version, Gemini 1.5, and outperforms it on a number of standard benchmarks, including MMLU-Pro, a large set of multiple-choice questions designed to test the capabilities of large language models in a range of subjects, from mathematics and physics to health, psychology and philosophy.

But the margins between top-end models like Gemini 2.0 and those from competing labs like OpenAI and Anthropic are now slim. These days, advances in large language models are less about how good they are and more about what you can do with them.

And this is where agents come into play.

Hands on with Project Astra

Last week I was led through an unmarked door on the top floor of a building in London’s King’s Cross into a room that strongly resembled a secret project. The word “ASTRA” was emblazoned in huge letters on one wall. Xu’s dog Charlie, the project’s de facto mascot, roamed between desks where researchers and engineers were busy developing a product on which Google is betting its future.