Google is in talks with chipmaker Marvell Technology to develop two new chips designed to run AI models more efficiently, according to a report from The Information on Sunday.
The report, citing two people familiar with the discussions, says one chip is a memory processing unit meant to work alongside Google's tensor processing unit (TPU). The other is a new TPU built specifically for running AI models.
This move is part of Google's broader push to make its TPUs a viable alternative to Nvidia's dominant GPUs. TPU sales have become a key driver of growth in Google's cloud revenue as the company tries to show investors its massive AI investments are paying off.
Reuters couldn't immediately verify the report, and neither Google nor Marvell responded to requests for comment.
If the talks pan out, the companies aim to finalize the design of the memory processing unit as soon as next year before handing it off for test production. This comes as Google is also exploring other AI hardware frontiers, like its partnership with Gucci on AI-powered smart glasses slated for 2027.