Inflection AI Building Exascale Supercomputing Cluster
Inflection AI has raised $1.3 billion from investors including Microsoft and Nvidia to build an AI supercomputing cluster using 22,000 of the latest Nvidia GPU chips, reports Nick Flaherty.
“Inflection AI is working with GPU cloud specialist CoreWeave on the cluster of H100 Tensor Core GPUs,” Flaherty says. “This provides 22 exaFLOPS of performance in the 16-bit precision mode, and even more if lower precision is used, and would be the world’s second most powerful supercomputer.”
Read more at EENewsEurope.
Related content
- Standing Up the El Capitan Exascale Supercomputer
- NVIDIA Announces Large Memory AI Supercomputer
-
The Road to End-of-Scale
The quest for exascale performance by the year 2020 is on. Delivering 1018 FLOPS could be slowed by the speed of light, among other things.
- ORNL’s Frontier Supercomputer Breaks the Exascale Barrier
- NERSC Gets Ready for Exascale
Subscribe to our ADMIN Newsletters
Subscribe to our Linux Newsletters
Find Linux and Open Source Jobs
Most Popular
Support Our Work
ADMIN content is made possible with support from readers like you. Please consider contributing when you've found an article to be beneficial.
