The book-sized device—which makes it the world’s smallest AI supercomputer—features Nvidia’s GB10 Grace Blackwell chip, 128GB unified memory, and up to 4TB storage.
It can run AI models with up to 200 billion parameters, while two connected units can handle Meta’s 405-billion-parameter Llama model.
“Placing an AI supercomputer on the desks of every data scientist, AI researcher, and student empowers them to engage and shape the age of AI,” said Nvidia CEO Jensen Huang at CES 2025 in Las Vegas.
The company also announced new software tools for building AI agents, including customized versions of Meta’s Llama model called Nemotron, optimized for autonomous task execution.
“In a lot of ways the IT department of every company is going to be the HR department of AI agents in the future,” Huang said.
Digits will be Nvidia’s most powerful consumer AI hardware, surpassing its existing Jetson lineup that starts at US$250. While capable of running sophisticated models, it won’t match the power of major cloud providers’ data centers.
This release could prove groundbreaking as it would represent a significant step toward democratizing access to powerful AI computing resources, though still at a premium price point.
