NVIDIA unveils Project DIGITS: Transforming AI supercomputing for developers worldwide

At CES 2025, NVIDIA introduced Project DIGITS, a revolutionary personal AI supercomputer designed to democratize access to high-performance AI tools. Equipped with the cutting-edge NVIDIA GB10 Grace Blackwell Superchip, this desktop-based system is set to redefine AI development by enabling users to prototype, fine-tune, and deploy large-scale models with ease.

By putting powerful AI capabilities at the fingertips of developers, data scientists, and students, NVIDIA aims to accelerate innovation across industries. Jensen Huang, NVIDIA’s CEO and founder, underscored the project’s significance, describing it as a step toward placing AI supercomputers on every developer’s desk, ensuring they are equipped to shape the future of AI-driven technology.

Revolutionary AI Performance with the GB10 Superchip

The NVIDIA GB10 Grace Blackwell Superchip lies at the heart of Project DIGITS, offering 1 petaflop of AI performance at FP4 precision. This level of computational power enables developers to experiment with cutting-edge models, including 200-billion-parameter language models, directly from their desktops.

Built on the NVIDIA Grace Blackwell architecture, the GB10 Superchip integrates an advanced Blackwell GPU featuring the latest CUDA cores and fifth-generation Tensor Cores. Its Grace CPU, designed with 20 Arm-based power-efficient cores, ensures unmatched performance and energy efficiency. NVIDIA collaborated with MediaTek to enhance the chip’s connectivity and design, making it one of the most efficient AI SoCs available.

The device also features 128GB of unified memory, up to 4TB of NVMe storage, and the ability to link two systems via NVIDIA ConnectX networking for handling models with up to 405 billion parameters.

From Local Prototyping to Scalable AI Deployment

Project DIGITS offers unmatched flexibility for AI development. Running on a Linux-based NVIDIA DGX OS, it allows developers to prototype and test AI models locally while offering seamless scalability to cloud or enterprise-grade infrastructures. Users can deploy their projects on NVIDIA DGX Cloud or data center systems that share the same Grace Blackwell architecture.

With access to the NVIDIA AI Enterprise software platform, developers can experiment with a wide range of tools and resources, including software development kits, orchestration tools, and pre-trained models from the NVIDIA NGC catalog. Popular frameworks like PyTorch, Python, and Jupyter Notebooks are fully supported, along with specialized tools like NVIDIA NeMo for fine-tuning models and NVIDIA RAPIDS for accelerating data science workflows.

The platform also provides advanced resources for building agentic AI applications, such as NVIDIA Blueprints and NIM microservices, which are available through the NVIDIA Developer Program. For production-level deployment, the AI Enterprise license offers enterprise-grade security, support, and software updates.

Driving Accessibility and Affordability in AI Supercomputing

Project DIGITS is not just about raw computing power—it is also designed for accessibility and affordability. Priced starting at $3,000, the system enables individual researchers, small teams, and educational institutions to access capabilities previously reserved for large enterprises.

The device operates with minimal energy requirements, running on a standard electrical outlet, further reducing barriers for widespread adoption. NVIDIA’s focus on scalability and seamless integration ensures that users can develop models locally and transition to more extensive infrastructures without losing continuity or compatibility.

Broader Implications for the AI Ecosystem

Industry experts view Project DIGITS as a pivotal moment in AI democratization. By bridging the gap between desktop AI development and large-scale deployment, NVIDIA is empowering smaller teams and individual innovators to compete with established players. This approach could accelerate advancements in fields like natural language processing, computer vision, and autonomous systems.

NVIDIA’s latest offering also aligns with broader industry trends, where the emphasis is shifting toward accessible and sustainable AI solutions. As the demand for efficient AI tools grows, Project DIGITS positions NVIDIA at the forefront of this evolving market.

Project DIGITS Availability and Future Outlook

NVIDIA has announced that Project DIGITS will be available in May 2025, both directly and through its top partners. Its competitive pricing, combined with unmatched performance, is expected to drive strong adoption across the developer community.

By providing a seamless pathway from prototyping to production, Project DIGITS underscores NVIDIA’s commitment to fostering innovation across the AI landscape. As the Grace Blackwell platform continues to evolve, NVIDIA is likely to further solidify its leadership in high-performance computing and AI development.