Condor Galaxy, the world’s “fastest AI training supercomputer”, has been built with assistance from G42, a UAE-based technology holding group. The machine is actually a network of nine interconnected AI supercomputers developed by US-based AI company Cerebras Systems.

Located in Santa Clara, California, the massive machine boasts 4 exaFLOPs of power and a staggering 54 million cores that will significantly reduce AI processing times.

G42 will use Condor Galaxy to train AI models across a variety of data sets and has already created and tested Arabic bilingual chat, healthcare, and climate study applications.

“Collaborating with Cerebras to rapidly deliver the world’s fastest AI training supercomputer and laying the foundation for interconnecting a constellation of these supercomputers across the world has been enormously exciting,” said Talal Alkaissi, CEO of G42 Cloud. “The partnership brings together Cerebras’ extraordinary compute capabilities, together with G42’s multi-industry AI expertise,” he added.

Training the latest AI models requires enormous computing power and specialized programming skills. ChatGPT, for example, relies on 175 billion parameters and uses 10,000 Nvidia GPUs to train its AI algorithms.

Condor Galaxy brings genuine innovation to these kinds of processes, as all computing is performed entirely without complex distributed programming languages. This means that large projects no longer require weeks or even months spent distributing work over thousands of GPUs.

Also Read: Best Web Hosting Providers In The Middle East

“Many cloud companies have announced massive GPU clusters that cost billions of dollars to build but are extremely difficult to use. Distributing a single model over thousands of tiny GPUs takes months from dozens of people with rare expertise,” noted Andrew Feldman, CEO of Cerebras Systems. “CG-1 eliminates this challenge. Setting up a generative AI model takes minutes, not months, and can be done by a single person” he added.

The G42 and Cerebras partnership marks another step toward the democratization of AI. The combination of massive computing power and unique AI data sets should produce groundbreaking results and turbocharge hundreds of AI projects around the world.