Google's Pixel 6 Will Boast 'Tensor' SoC Built for ML and AI

Google today announced that it designed a custom system on a chip (SoC) to improve the machine learning and artificial intelligence performance of its next smartphones. The new chip, which the company dubbed Tensor, is set to debut in the Pixel 6 and Pixel 6 Pro "later this fall" alongside the Android 12 operating system.

The company uses the Tensor identifier for many of its ML- and AI-related projects. Its open source machine learning platform is called TensorFlow. The application-specific integrated circuits (ASICs) on which Google Cloud relies are called Tensor Processing Units. And now it's built a custom Tensor SoC for its own smartphones.

Unfortunately the announcement was light on details — Google didn't offer any information about the Tensor SoC's tech specs or provide a specific release date for the Pixel 6 lineup. But it did offer some insight into what it set out to accomplish with its first mobile SoC and what that could mean for Android users once it debuts.

"AI is the future of our innovation work," Google said, "but the problem is we’ve run into computing limitations that prevented us from fully pursuing our mission. So we set about building a technology platform built for mobile that enabled us to bring our most innovative AI and machine learning (ML) to our Pixel users."

"You’ll see this in everything from the completely revamped camera system to speech recognition and much more," Google said. "So whether you're trying to capture that family photo when your kids won’t stand still, or communicate with a relative in another language, Pixel will be there — and it will be more helpful than ever."

Nathaniel Mott
Freelance News & Features Writer

Nathaniel Mott is a freelance news and features writer for Tom's Hardware US, covering breaking news, security, and the silliest aspects of the tech industry.

  • Vyrvelata
    I can't wait to see some benchmarks... my hopes are to be close with Qualcomm 865 but.. I'm fine with less also.... but what if it sweeps the floor of all android arm CPU's (highly unlikely.. they are going for better AI integration instead).
    Reply