At CES 2025 in Las Vegas, Nvidia unveiled Mission Digits, a “private AI supercomputer” that gives entry to the corporate’s Grace Blackwell {hardware} platform in a compact type issue.
“[Project Digits] runs your entire Nvidia AI stack — all of Nvidia software program runs on this,” Nvidia CEO Jensen Huang stated onstage throughout a press convention on Monday. “It’s a cloud computing platform that sits in your desk … It’s even a workstation, should you prefer it to be.”
Designed for AI researchers, knowledge scientists, and college students, Mission Digits packs Nvidia’s new GB10 Grace Blackwell Superchip, which delivers as much as a petaflop of computing efficiency for prototyping, fine-tuning, and operating AI fashions.
Nvidia claims a single Mission Digits unit can run fashions as much as 200 billion parameters in measurement. Parameters roughly correspond to a mannequin’s problem-solving expertise, and fashions with extra parameters usually carry out higher than these with fewer parameters.
The GB10, which was developed in partnership with MediaTek, options an Nvidia Blackwell GPU linked to a 20-core Nvidia Grace CPU. Contained in the Mission Digits enclosure, the chips are hooked as much as a 128GB pool of reminiscence and as much as 4TB of flash storage.
Nvidia says that two Mission Digits machines will be linked collectively to run as much as 405-billion-parameter fashions, if a job requires it. Mission Digits can ship a standalone expertise, as alluded to earlier, or hook up with a major Home windows or Mac PC.
But it surely’s not low-cost. Mission Digits machines, which run Nvidia’s Linux-based DGX OS, will probably be out there beginning in Might from “prime companions” for $3,000, the corporate stated.
So it gained’t be simply anybody who can afford a Mission Digits unit of their very own. Huang thinks that there’s a market, nevertheless.
“With Mission Digits, the Grace Blackwell Superchip involves tens of millions of builders,” he stated in a press release. “Putting an AI supercomputer on the desks of each knowledge scientist, AI researcher and scholar empowers them to interact and form the age of AI.”