NVIDIA has rolled out Project Digits, a personal AI supercomputer that brings unprecedented computing power to the desks of researchers, data scientists, and students worldwide.
Powering Project Digits is the NVIDIA GB10 Grace Blackwell Superchip that delivers up to one petaflop of AI performance at FP4 precision. This system-on-a-chip (SoC) combines NVIDIA’s latest Blackwell GPU technology with a high-performance Grace CPU, featuring 20 power-efficient Arm-based cores.
With 128GB of unified, coherent memory and up to 4TB of NVMe storage, Project Digits enables developers to run AI models with up to 200 billion parameters directly on their desktops — a level level of performance that was previously only available in large-scale data centres or through cloud services.
NVIDIA’s strategic approach extends beyond raw computing power. Project Digits runs on the Linux-based NVIDIA DGX OS, allowing seamless deployment of models to NVIDIA DGX Cloud, accelerated cloud instances, or data centre infrastructure. Developers can now prototype on their desks and scale effortlessly to larger computing environments.
Accelerating AI Innovation
“With Project Digits, the Grace Blackwell Superchip comes to millions of developers. 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 Jensen Huang, Founder and CEO of NVIDIA.
Positioned as a complete AI development environment right out of the box, the system comes with access to NVIDIA’s extensive AI software library, including development kits, frameworks, and pre-trained models.
Perhaps most surprising is the pricing that starts from US$3,000, placing Project Digits well within reach of a wide range of users. The system will be available from NVIDIA and selected partners beginning in May 2025.
Entelechy Asia 