Nvidia unveils Project Digits, an AI-powered personal supercomputer for researchers and students

Nvidia Today Nvidia revealed Project numbersan artificially intelligent super-personal computer that provides researchers, data scientists and students around the world with access to the power of Nvidia Grace Blackwell
platform.

Project Digits features the new Nvidia GB10 Grace Blackwell Superchip processor, which delivers petaflops of AI computing performance for prototyping, tuning, and running large AI models. The company made the announcement during the opening speech by CEO Jensen Huang Fei Consumer Electronics Show 2025the big tech trade show in Las Vegas this week.

With Project Digits, users can develop and run model inference using their desktop system, then seamlessly deploy the models to the cloud or accelerating data center infrastructure. Huang said it relies on a “top-secret chip called GB110, which is the smallest chip we can make from Blackwell.”

“AI will be prevalent in every application for every industry. With Project Digits, the Grace Blackwell Superchip reaches millions of developers.” “Putting an AI supercomputer on the desks of every data scientist, AI researcher and student empowers them to participate and shape
The era of artificial intelligence.

GB10 Superchip delivers petaflops of power-efficient AI performance The GB10 Superchip is a system-on-chip (SoC) based on the Nvidia Grace Blackwell architecture that delivers up to 1 petaflop of AI performance at FP4 resolution.

The GB10 features an Nvidia Blackwell GPU with the latest generation CUDA cores and a 5th generation Tensor core, connected via a chip-to-chip NVLink-C2C link to a high-performance Nvidia Grace CPU, which includes 20 power-efficient cores engineered using the ARM architecture. MediaTek, the market leader in Arm-based SoC designs, collaborated on the design of the GB10, delivering best-in-class power, 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 unified, coherent memory and up to 4TB of NVMe storage. Using the supercomputer, developers can run up to 200 billion large language models to power AI innovations. Additionally, using the Nvidia ConnectX network, two Project DIGITS AI supercomputers can be linked to run up to 405 billion parameter models.

Grace Blackwell AI supercomputing at hand

Using the Grace Blackwell architecture, organizations and researchers can prototype, fine-tune, and test models on on-premises Project DIGITS systems running Linux-based Nvidia DGX OS, and then seamlessly deploy them on Nvidia DGX Cloud, accelerated cloud instances, or data center infrastructure.

This allows developers to build an AI prototype on Project DIGITS and then scale to 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 experimentation and prototyping, including SDKs, orchestration tools, frameworks, and models available in the Nvidia NGC catalog and on the Nvidia Developer Portal. Developers can fine-tune forms using
The Nvidia NeMo framework accelerates data science by using Nvidia Rapids libraries and running popular frameworks such as PyTorch, Python, and Jupyter notebooks.

To build powerful AI applications, users can also harness Nvidia Blueprints and Nvidia NIM microservices, available for R&D and testing via the Nvidia Developer Program. When AI applications are ready to move from experimental to production environments, Nvidia AI Enterprise licensing provides security, support, and product releases for enterprise-level Nvidia AI software.

Availability

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

“By making Llama models open source, we are committed to democratizing access to cutting-edge AI technology. Using Project Digits, developers can harness the power of llamas locally, opening up new possibilities for innovation,” Ahmed Al-Dahla, head of GenAI at Meta, said in a statement. And cooperation.”

“Developing artificial intelligence requires tools that enable researchers to conduct experiments at scale, quickly and accurately. Nvidia’s Project Digits represents a major leap forward. I’m excited to see how 128GB in this “Small size will enhance the future of enterprise AI.”

“At Hugging Face, we want to make it easier for developers to build their own AI. Nvidia’s Digits project will enable AI builders to build and run their own AI models and systems at the edge. With 128GB of unified memory, AI builders can “I can’t wait to see what he builds,” Je Boudier, head of product at Hugging Face, said in a statement Hugging Face Community Using Nvidia Project Digits”.

“Nvidia’s Project Digits is a powerful resource that you can hold in the palm of your hand. With two Project Digits modules, developers can easily work with AI models up to 405 billion parameters in size. We can’t Waiting to see what applications people will build using this.”

Daily insights into business use cases with VB Daily

If you want to impress your boss, VB Daily has you covered. We give you the inside scoop on what companies are doing with generative AI, from organizational transformations to hands-on deployments, so you can share insights to maximize ROI.

Read our privacy policy

Thanks for subscribing. Check more VB newsletters here.

An error occurred.