Computer

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

admin
admin

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

Source: Tom's Hardware

You Might Also Like

Tower Semiconductor to invest $4 billion in Japanese ops to set up massive optical connectivity hub
Micron discontinues 2GB GDDR7 chips for gaming GPUs as it pivots toward higher-density memory for AI
ASML snubs Elon Musk-backed particle accelerator chipmaking tech — firm doubles down on 1,000W laser-produced plasma systems for chipmaking tools
Qualcomm will pay Huawei to license its patents for the first time in historic turnaround
Redefining enterprise intelligence with autonomous AI
TAGGED:
Share This Article
Previous Article Former U.S. Ambassador to Ukraine William Taylor on Trump's embrace of Russia
Next Article Hugging Face’s New AI Biology Model Lets You Chat With DNA
Leave a Comment

Leave a Reply

Your email address will not be published. Required fields are marked *