NVIDIA has made a significant announcement at CES 2025, introducing Project DIGITS, a groundbreaking personal AI supercomputer designed to provide AI researchers, data scientists, and students worldwide with access to the powerful NVIDIA Grace Blackwell platform.
This innovative offering features the new NVIDIA GB10 Grace Blackwell Superchip, capable of delivering up to a petaflop of AI computing performance, effectively empowering users to prototype, fine-tune, and run large AI models right from their desktops.
Empowering the AI Community
Jensen Huang, NVIDIA’s founder and CEO, emphasized the transformative impact of Project DIGITS, stating, “AI will be mainstream in every application for every industry. With Project DIGITS, the Grace Blackwell Superchip comes to millions of developers.”
This initiative aims to democratize AI technology, placing powerful computing capabilities in the hands of AI professionals and enthusiasts alike.
GB10 Superchip: Unmatched Performance
The heart of Project DIGITS is the GB10 Superchip, an advanced system-on-a-chip (SoC) that utilizes the NVIDIA Grace Blackwell architecture.
The GB10 delivers up to 1 petaflop of AI performance at FP4 precision, combining an NVIDIA Blackwell GPU equipped with cutting-edge CUDA cores and fifth-generation Tensor Cores, alongside a high-performance NVIDIA Grace CPU featuring 20 energy-efficient cores designed on the Arm architecture.
Notably, MediaTek played a key role in developing the GB10, enhancing its power efficiency and connectivity.
With its efficient design, Project DIGITS can operate with a standard electrical outlet, featuring 128GB of unified memory and up to 4TB of NVMe storage.
This setup allows developers to run massive 200-billion-parameter large language models, further fueling AI innovation.
Moreover, users have the option to link two Project DIGITS systems, enabling them to tackle models with up to 405 billion parameters.
Seamless Integration and Scalable Solutions
The Grace Blackwell architecture supports a streamlined workflow for enterprises and researchers alike.
Users can prototype and test models locally on the Project DIGITS systems, which operate on the Linux-based NVIDIA DGX OS, and seamlessly deploy their models to NVIDIA DGX Cloud™ or other accelerated cloud and data center infrastructures.
This capability ensures consistency in the development environment while scaling up as needed.
Project DIGITS also opens the door to an extensive library of NVIDIA AI software, available for experimentation and prototyping.
This includes a variety of software development kits, orchestration tools, frameworks, and models accessible through the NVIDIA NGC catalog and the NVIDIA Developer portal.
Developers can leverage tools like the NVIDIA NeMo™ framework for fine-tuning models, enhance data science workflows with NVIDIA RAPIDS™ libraries, and utilize popular frameworks such as PyTorch, Python, and Jupyter notebooks.
To aid in the creation of advanced AI applications, users can tap into NVIDIA Blueprints and NVIDIA NIM™ microservices, which are available through the NVIDIA Developer Program.
For those looking to transition their applications from experimentation to production, the NVIDIA AI Enterprise license provides robust security, support, and access to the latest NVIDIA AI software releases.
Availability and Pricing
According to the NVIDIA Report, Project DIGITS is set to launch in May 2025, with pricing starting at $3,000. Interested parties can sign up for notifications to learn more about this exciting new technology.
As NVIDIA continues to push the boundaries of AI computing, Project DIGITS stands as a testament to their commitment to making powerful AI tools accessible to a broader audience, paving the way for innovation in the field of artificial intelligence.
%20(1).webp?w=1200&resize=1200,0&ssl=1 "Data Breach at Legends International Exposes Customer Information")
%20(1).webp?w=1200&resize=1200,0&ssl=1 "Critical AnythingLLM Vulnerability Enables Remote Code Execution")
 that utilizes the NVIDIA Grace Blackwell architecture.){kind=link}