(via techspot)
The new CS-1 can provide 15kw power for wafer level chips (1kW lost due to low power efficiency).
The cerebras wafer scale engine it supports is 56 times the largest GPU ever, 78 times the core, 3000 times the on-chip memory, and 33000 times the bandwidth.
In other words, CS-1 can provide an extremely fast computing experience. In addition, it can be used with open source machine learning frameworks (such as pytorch and tensorflow) to improve application flexibility.
As for more details of the hardware (such as clock speed), the company said it would share with you in the near future. (the size of a wafer level chip is equivalent to an iPad tablet.)
The price of such an amazing set of equipment is obviously quite high. A spokesman for the company said the cost was in the millions of dollars.
But for customers who can afford it, it's not a big deal. For example, Argonne National Laboratory has one, which focuses on basic scientific experiments such as drugs that can be used for cancer research and treatment.