2 min

Tags in this article

, ,

Google’s Coral chips are suitable for adding features that use machine learning to edge devices. The product line has a number of new models based on the Edge TPU.

This Edge TPU is a smaller version of the Tensor Processing Unit, a Google processor optimised for AI. The smaller version has traded power for efficiency, which is why it can be so small.

The Accelerator Module is the first new version of the chip, and comes in a bare-bones housing that can be mounted in four different ways in a gadget. Among other things, PCIe and USB connections are possible.

In addition, two new versions of Coral System-on-Module have been announced. These modules also have an Edge TPU, but also a CPU, graphics chip, memory and pre-installed Linux. This product should bring AI to the user faster than the above mentioned bare-bones Edge TPU. The two different versions have either 2 or 4 gigabytes of LPDDR4 memory.

Dev Board Mini

Then there’s Dev Board Mini, a smaller version of Dev Board. This solution is intended for development teams to test machine learning functions. The platform has an Edge TPU, a CPU, a GPU, 2 gigabytes of RAM, 8 gigabytes of flash memory and also a built-in Linux OS.

According to Google, for example, the new Edge TPU can run multiple computer vision models at 30 fps, or one model at a time at around 400 fps.

“More and more industries are beginning to recognize the value of local AI, where the speed of local inference allows considerable savings on bandwidth and cloud compute costs, and keeping data local preserves user privacy,” Coral Director of Research Billy Rutledge stated. He added that companies have in the past year harnessed the modules for applications “across a broad set of industries that range from healthcare to agriculture to smart cities.”