Nvidia unveils Project Digits personal AI supercomputer for researchers and students

Nvidia Nvidia announced today Digital projecta personal AI supercomputer that gives AI researchers, data scientists and students around the world access to the power of the Nvidia Grace Blackwell
platform.

Project Digits introduces the new Nvidia GB10 Grace Blackwell Superchip, offering a petaflop of AI computing performance for prototyping, tuning and running large AI models. The company made the announcement during CEO Jensen Huang's keynote at CES 2025the big tech trade show in Las Vegas this week.

With Project Digits, users can develop and run a decision using their own desktop system, and then quickly deploy the models on cloud infrastructure or an accelerated data center. It's based on a “super secret chip called GB110, the smallest Blackwell we can make,” Huang said.

“AI will become mainstream in every application for every industry. With Project Digits, the Grace Blackwell Superchip will come to millions of developers,” said Huang. “Putting an AI supercomputer on the desks of every data scientist empowers AI researchers and students to engage and shape the
the age of AI.”

GB10 Superchip is a system-on-a-chip (SoC) based on Nvidia's Grace Blackwell architecture and delivers up to 1 petaflop of AI performance at FP4 precision.

GB10 features an Nvidia Blackwell GPU with latest-generation CUDA cores and fifth-generation Tensor Cores, connected via NVLink-C2C chip-to-chip interconnect to a high-performance Nvidia Grace CPU, which provides a including 20 energy efficient cores built by the Arm. architecture. MediaTek, the market leader in Arm-based SoC designs, collaborated on the GB10 design, adding best-in-class power efficiency, performance and connectivity.

The GB10 Superchip enables Project DIGITS to deliver powerful performance using only a standard electrical outlet. Each DIGITS Project features 128GB of integrated, coherent memory and up to 4TB of NVMe storage. With the supercomputer, developers can run up to 200-billion-parameter large language models to apply AI innovation. In addition, using Nvidia ConnectX networking, two Project DIGITS AI supercomputers can be connected to run up to 405-billion-parameter models.

Grace Blackwell AI Supercomputing within reach

With Grace Blackwell's architecture, enterprises and researchers can prototype, fine-tune and test models on local Project DIGITS systems running Linux-based Nvidia DGX OS, and then quickly deploy them on Nvidia DGX Cloud, environments accelerated cloud or data center infrastructure.

This allows developers to prototype AI on Project DIGITS and then scale to a cloud or data center infrastructure, using the same Grace Blackwell architecture and Nvidia AI Enterprise software platform.

Project Digits users can access an extensive library of Nvidia AI software for testing and prototyping, including software development tools, orchestration tools, frameworks and modules available in the Nvidia NGC catalog and on Nvidia Developer portal. Developers can tune modules with the
Nvidia NeMo framework, accelerate data science with Nvidia Rapids libraries and run common frameworks like PyTorch, Python and Jupyter notebooks.

To build agent AI applications, users can also make use of Nvidia Blueprints and Nvidia NIM microservices, which are available for research, development and testing through the Nvidia Development Program. When AI applications are ready to move from testing to production environments, the Nvidia AI Enterprise license provides enterprise-grade security, support and product releases of Nvidia AI software.

Available

Project Digits will be available in May from Nvidia and key partners, starting at $3,000.

“By making Llama models open source, we are committed to democratizing access to cutting-edge AI technology. With Project Digits, developers can harness the power of Llama locally, unlocking new opportunities for innovation and collaboration,” said Ahmad Al-Dahle, Head of GenAI at Meta, in a statement.

“Advancing AI requires tools that empower researchers to conduct experiments at scale, speed and precision. Nvidia's Digits Project represents a major breakthrough. I'm excited to see how 128GB in such a small form factor can advance enterprise AI,” said Silvio Savarese, Chief Scientist at Salesforce, in a statement.

“At Hugging Face, we want to make it easy for developers to build their own AI. Nvidia's Digits Project empowers AI builders to build and run their own Gen AI models and systems at the edge. With 128GB of integrated memory, AI builders can run 200B parameter models locally, and connect multiple Project Digits systems to scale from there. I can't wait to see what the Hugging Face community builds with Nvidia Project Digits,” said Je Boudier, Head of Product at Hugging Face, in a statement.

“Nvidia's Digits Project is a powerhouse you can hold in the palm of your hand. With two Project Digits units, developers can easily work with AI models up to 405B parameters in size. We can't wait to see what apps people build with this,” Michael Chiang, co-founder of Ollama, said in a statement.