Simply hours after Dell determined all by itself to rename its laptop computer line with Professional and Professional Max monikers, Nvidia additionally took a web page from Apple’s playbook with its new “private AI supercomputer.” No, it’s not referred to as the Nvidia mini, however it does have a sure simpatico with the Mac mini.
Nvidia didn’t share the precise specs for what it’s calling Mission Digits, however the system match neatly in CEO Jensen Huang’s palm throughout its unveiling. Subsequent to a monitor and keyboard (above), it seems to be barely larger than the Mac mini with a boxier and extra industrial design.
Nvidia didn’t present benchmarks for the GB10 chip, however did say it’s able to a petaflop of FP4 floating level efficiency and capable of “run as much as 200-billion-parameter massive language fashions,” possible past the capabilities of the $599 Mac mini. Nonetheless, Nvidia isn’t promoting it to simply anybody. The objective of Mission Digits, Huang stated, is to place “an AI supercomputer on the desks of each information scientist, AI researcher and pupil.”
Nvidia
To that finish, the pc begins at $3,000 and comes with a 20-core GB10 Grace Blackwell Superchip and 128GB of unified reminiscence. Configuring a Mac mini with the M4 Professional chip, which has a 14-core CPU, and 64GB of RAM, prices $2,499 with 1TB of storage. A 1TB Mac Studio fitted with a 24-core M2 Extremely and 128GB of RAM brings the value to $4,799.
Whereas it takes a number of M4 Mac minis to even attain a teraflop of efficiency, the M2 Extremely within the Mac Studio already tops 20 teraflops. And Apple is predicted to improve that machine to the M4 household of processors someday this yr, which can clearly deliver a significant efficiency enhance. And we assume the ability draw can be many instances that of the Mac mini or Mac Studio (or Mac Professional for that matter).
Nvidia
After all, we’re evaluating Apples to oranges right here. Nvidia’s Mac mini clone isn’t meant for on a regular basis work. Digits is kitted out with Nvidia’s full AI software program enterprise platform, together with “an intensive library of Nvidia AI software program for experimentation and prototyping, together with software program growth kits, orchestration instruments, frameworks and fashions.”
