Alphabet’s Google is in talks with Marvell Technology to develop two new chips aimed at running AI models more efficiently.
The report says that one of the chips is a memory processing unit designed to work with Google’s Tensor Processing Unit (TPU) and the other chip is a new TPU specifically designed to run AI models.
Information Google is pushing to make its TPU a viable alternative to Nvidia’s flagship GPU, reports said on Sunday. TPU sales have become a key driver of Google’s cloud revenue growth as it aims to show investors that its AI investments are generating returns.
According to the report, the companies aim to finalize the design of the memory processing unit by next year before handing it over for trial production.
The latest news comes as Google recently announced a deal with the Pentagon, which will help Alphabet expand its government relationships as the US takes steps to incorporate artificial intelligence into its operations to reduce costs and speed up administrative tasks.
The Alphabet-owned platform reported Thursday that it is negotiating a deal with the Defense Department that would allow the Pentagon to deploy its Gemini AI model in classified settings.
