Intel unveils ‘world’s largest AI computer that mimics human brain’ with 1,000 chips – it’s ’50x faster than rivals’
INTEL has built the world’s largest neuromorphic computer that aims to function similarly to the human brain.
The tech giant announced the computer, dubbed Hala Point, on Wednesday, April 17.
Hala Point comprises 1,152 Loihi 2 chips that mimic a total of 1.15 billion artificial neurons.
Intel’s Loihi 2 processor hopes to support future brain-inspired artificial intelligence research.
There are also an impressive “128 billion synapses distributed over 140,544 neuromorphic processing cores,” Intel revealed.
The computer advances on Intel’s first-generation research system, Pohoiki Springs, the company said.
This includes architectural improvements that achieve over 10 times more neuron capacity and up to 12 times higher performance.
It can also make 20 quadrillion operations per second, Live Science reported.
Mike Davies, director of the Neuromorphic Computing Lab at Intel Labs, said the computing cost of today’s AI models is growing at unsustainable rates.
As such, the industry needs new approaches and for that reason, Intel developed Hala Point.
Hala Point combines deep learning efficiency with brain-inspired learning and optimization capabilities.
“We hope that research with Hala Point will advance the efficiency and adaptability of large-scale AI technology,” Davies added.
Sandia Labs, which worked with Intel to develop Hala Point, hopes to use the system for “brain-scale computing research.”
Specifically, it wants to target areas like device physics, computer architecture, computer science, and informatics.
“Working with Hala Point improves our Sandia team’s capability to solve computational and scientific modeling problems,” said Craig Vineyard, Hala Point team lead at Sandia National Laboratories.
“Conducting research with a system of this size will allow us to keep pace with AI’s evolution in fields ranging from commercial to defense to basic science,” he added.
Currently, Hala Point is only a research prototype, but it could soon be used in commercial systems.
“For the first time we’re showing standard deep neural networks being mapped and transformed into a form that can run at this kind of scale in a neuromorphic system,” Davies told ZDNET.
“That is a first for anybody, to show that standard deep neural networks can actually, with some caveats, run with the competitive efficiency on par with the very best GPUs and, ASICs [application-specific integrated circuits] that are being produced now.”
Intel is an American multinational corporation and technology company headquartered in Santa Clara, California.
The company was founded in 1968 and is one of the world’s largest semiconductor chip manufacturers.