Meta, formerly known as Facebook, has chosen Microsoft Azure as a strategic cloud provider for AI research and development. The news was announced during Build 2022, as spotted by ZDNet, and is the latest step in the two company's partnership in this area.
Under the partnership, Meta will expand the use of supercomputing power, using a dedicated Azure cluster of 5,400 GPUs. That includes the latest VM series on Azure, which has Nvidia A100 Tensor Core 80GB GPUs. According to Microsoft, Meta used Azure services in the past to train their recent OPT-175B language model.
Additional parts of this partnership will see Microsoft and Meta collaborate on scaling PyTorch adoption on Azure and accelerate developers' journey from experimentation to production.
"We are excited to deepen our collaboration with Azure to advance Meta’s AI research, innovation, and open-source efforts in a way that benefits more developers around the world,” said Jerome Pesenti, VP of AI, Meta.
With Build 2022 still underway, you can expect a lot more announcements like this one. Artificial Intelligence is a hot topic for Microsoft, particularly on the development front, as best seen with Project Volterra, Power Pages, and more.