Nvidia’s Project DIGITS: A Personal AI Supercomputer Redefining Accessibility and Power for AI Development Worldwide.

In his keynote speech at CES 2025, held in Las Vegas on Monday, NVIDIA’s CEO Jensen Huang made a lot of announcements where the focus was primarily on AI. Nvidia’s CEO has launched the latest GeForce RTX 50 series GPUs, powered by the new Blackwell architecture, the Cosmos AI model-that will train robots and self-driving cars, a new set of open-source large language models (LLMs), and many more developments around AI.

The highlight of the announcement was the introduction of Project DIGITS, Nvidia’s first personal AI supercomputer, that makes high-performance AI developments accessible to developers, researchers, and enterprises.The $3000 Project Digits, powered by the NVIDIA GB10 Grace Blackwell Superchip, integrates advanced processing capabilities for AI workloads. It runs on a Linux-based operating system. It is wireless and can be connected to the desktop, user can develop and run inference on models, and then seamlessly deploy the models on accelerated cloud or data center infrastructure. The supercomputer will be available by May 2025.

Key Features of Nvidia’s Project DIGITS:

  • Powerful Hardware:

It features the GB10 Grace Blackwell Superchip, which integrates the Grace CPU with 20 arm cores (10 Cortex-X925 and 10 Cortex-A725), and the Blackwell GPU, capable of delivering up to 1 petaflop (1,000 TFLOPs) of AI computing power at FP4 precision. Equipped with Nvidia’s latest CUDA and RT cores the Blackwell GPU enables users to handle and fine-tune large AI models with up to 200 billion parameters of size.

  • Unified Memory:

Project Digits is equipped with 128 GB of unified memory and 4TB of flash storage, which facilitates seamless processing of large datasets, and complex AI models can be processed without any bottlenecks commonly associated with traditional systems.

  • Extensive Software Support:

Project DIGITS runs the entire Nvidia AI stack, including all its libraries and frameworks. Developers can utilize tools like the NVIDIA NeMo framework for model fine-tuning and RAPIDS libraries for data science acceleration.

Traditionally, AI development has relied heavily on cloud-based solutions or costly data centers. Project DIGITS is changing this paradigm by offering a powerful desktop solution that democratizes access to high-performance AI tools. It allows users to develop complex AI models locally while retaining the option to scale their projects using cloud resources. The supercomputer available along with the desktop will empower AI developers and researchers, Enterprises and Universities, and tech innovators reshape the age of AI.