Alphabet Inc.’s subsidiary Google is reportedly in discussions with Marvell Technology to co-develop two new chips aimed at improving the efficiency of artificial intelligence (AI) models, according to a media report.
Manchester City Edge Arsenal in Thriller to Shake Up Title Race
The proposed collaboration includes a memory processing unit designed to work alongside Google’s existing Tensor Processing Units (TPUs), as well as a next-generation TPU specifically optimized for running AI workloads more efficiently.
Google has been accelerating efforts to position its TPUs as a strong alternative to the GPUs produced by Nvidia, which currently dominate the AI hardware market. The company’s in-house chips have already become a significant contributor to its cloud computing revenue, particularly as demand for AI infrastructure continues to surge.
According to the report, the companies are aiming to finalize the design of the memory processing unit as early as next year, after which it would move into test production.
The development reflects Google’s broader strategy to strengthen its AI ecosystem and demonstrate tangible returns on its substantial investments in artificial intelligence technologies. However, the report has not yet been independently verified, and both companies have not issued official comments on the matter.
