Google’s TPUs for AI workloads now run in Meta's data centers

T2Covered by 1 source · first reported by Tom's Hardware at 22 Oct, 10:00 UTC

Google’s TPUs now serve AI in production at Meta, through new open-source engines that run PyTorch models natively on the chips.

clouddevhardware

Sources

• T2Google’s TPUs for AI workloads now run in Meta's data centersTom's Hardware22 Oct, 10:00