by Andrew Cochran

A California start-up, Cerebras, has unveiled the world’s largest computer chip. Reports say the chip can process 4 billion parameters simultaneously, opening the door to a new scale of AI computing.

The Cerebras chip is the world’s largest, 56x bigger than any before and with 1.2 trillion transistors.

The 4-billion parameter capacity of the new chip compares to about a million per chip typically in processors today.

Cerebras calls their new chip the WSE, which stands for ‘wafer scale engine.’ Forbes calls it ‘a supercomputer-on-a-chip.’

The breakthrough is cutting out the interconnections. Chips often are ganged together to deliver faster outputs, but each connection involves time to trade results. Putting, say, ten chips together does not offer 10x performance.

BBC News says the company expects processing time for some complex operations can drop ‘from months to minutes.’

ZDNET says WSE will enable new kinds of neural networks, and EE|Times reports the WSE offers similar performance to an assembly of 1,000 chips by its closest competitor ‘while requiring just 2-3% of its space and power.’

Wired gives an example of $350,000 in computing time for a single piece of language software. Dropping computing time can save costs, which can be in the millions when working with large AI models.

The price for the WSE and its housing has not been reported.

SEE RELATED STORIES

LATEST