A device that will reduce AI energy consumption by 1000 times has been developed!
says Yang Lv lead author of the study
With CRAM, data is processed directly in memory cells, resulting in incredible energy savings.
This new hardware device CRAM Computational Random Access Memory, can reduce energy consumption for AI applications by a full 1000 times
It is estimated that a CRAM-based machine learning accelerator can, save up to 2500 times more energy compared to traditional methods
