Nvidia is set to launch the DGX Spark ‘Personal AI Supercomputer’ on October 15.

This compact device is made to sit on your desktop, allowing users to develop advanced Artificial Intelligence (AI) models.

Starting this Wednesday, you can buy the DGX Spark.

Nvidia dubs it “the world’s smallest AI supercomputer”.

Nvidia’s CEO, Jensen Huang, mentioned that “having an AI supercomputer available for every data scientist, AI researcher, and student allows them to participate in and influence the AI era”.

Huang first introduced Spark, which was previously called Digits, earlier this year.

Spark delivers performance that used to be possible only with expensive, high energy data centers.

Nvidia has mentioned that third party manufacturers can make their own versions of Spark, we can expect to see some similar devices hitting the market.

Nvidia has announced that Acer, Asus, Dell, Gigabyte, HP, Lenovo, and MSI will all be launching their own tailored versions of Spark.

Spark boasts up to 4TB of SSD storage, 128GB of unified memory, and features Nvidia’s GB10 Grace Blackwell Superchip.

It can manage AI models with as many as 200 billion parameters and deliver a petaflop of AI performance, which is a million billion calculations every second.

Plus, it’s compact, fits perfectly on a desk, and can be plugged into a standard outlet.

People can find it online at nvidia.com and from select partners and retailers throughout the United States.

Nvidia has revealed that the DGX Spark will cost $3,999 (about Rs 3,55,106).

Most PC manufacturers have their own customized supercomputer models, and the Acer Veriton GN100 is a similar option available for the same price of $3,999.

DISCLAIMER: The specifications and features listed are derived from official announcements and might differ based on the region or model. Be sure to check the details on the official website or with authorized sellers before you buy.

Subscribe My Channel





Discover more from Connect2ConnectOnline

Subscribe now to keep reading and get access to the full archive.

Continue reading