At Google I/O, the company shared their next generation AI processing chip, the Tensor Processing Unit (TPU) v4.

Machine learning has become critically important in recent years, powering critical services like the Google Assistant and really almost app and service that Google offers. To ensure AI and machine learning powered experiences run efficiently, Google has developed the dedicated Tensor Processing Unit.

During Google I/O 2021, Sundar Pichai unveiled the fourth-generation Tensor Processing Unit. A typical pod of 4096 TPUs is reportedly capable of speeds of over 1 exaflop.

Pods of the Tensor Processing Unit v4 are already being deployed in Google’s datacenters and later this year, they will be made available to Google Cloud customers.


FTC: We use income earning auto affiliate links. More.

Check out 9to5Google on YouTube for more news:

You’re reading 9to5Google — experts who break news about Google and its surrounding ecosystem, day after day. Be sure to check out our homepage for all the latest news, and follow 9to5Google on Twitter, Facebook, and LinkedIn to stay in the loop. Don’t know where to start? Check out our exclusive stories, reviews, how-tos, and subscribe to our YouTube channel

About the Author

Kyle Bradshaw

Kyle is an author and researcher for 9to5Google, with special interests in Made by Google products, Fuchsia, and Stadia.

Got a tip or want to chat? Twitter or Email.

Kyle Bradshaw's favorite gear