The Facebook company confirmed reports that it is developing the microchips, EEtimes reported.
Facebook “is absolutely bringing up a silicon team focused on working with silicon providers, and we have a chip we’re building, but it’s not our primary focus,” said vice president of infrastructure at Facebook, Jason Taylor.
According to him, the chip is “not the equivalent of [Google’s] TPU” deep-learning accelerator. The information was revealed at the @Scale event, where Facebook said that five chip companies would support Glow, an open-source, deep-learning compiler that it backs.
The Glow is expected to allow developers to use any of the new deep learning accelerators in applications for the cloud or edge of the network. It does not target client systems such as smartphones.
“We expect that there will be hardware fragmentation [in inference accelerators]. Our work with Glow is to help machine-learning experts design neural nets and not have to do the work required to tune them to each unique chip,” Taylor added.