Nvidia unveils DGX Spark (rebranding DIGITS) and launches Blackwell DGX Station for desktop AI supercomputing.
NVIDIA has introduced two new personal AI supercomputers, the DGX Spark and DGX Station, powered by the Grace Blackwell platform and designed for AI developers, researchers, data scientists, and students who require powerful computing capabilities for prototyping, fine-tuning, and running AI models. The DGX Spark is priced at USD 3,000, while the DGX Station's price has not yet been announced, though previous iterations of the DGX Station have started from USD 99,000. These systems represent NVIDIA's effort to make high-performance AI computing more accessible to a broader audience.
The DGX Spark is, in reality, a rebranded and updated version of Project DIGITS, originally launched in 2015 as an AI development platform aimed at deep learning practitioners. Built on top of CUDA and cuDNN, DIGITS was essentially middleware for deep learning GPU training that would be packaged with DRIVE as part of NVIDIA's training solution for designing self-driving cars. DIGITS was first introduced in March 2015 at NVIDIA's GPU Technology Conference and has been iterated upon since then. Earlier this year at CES, NVIDIA announced that it would be packaging DIGITS with a GPU as an out-of-the-box solution for anyone to build a neural network training system. With the introduction of the Blackwell GPU architecture, NVIDIA has now repositioned this product under the DGX branding, emphasizing its suitability for AI-native applications.
The DGX Spark is positioned as the smallest AI supercomputer, bringing powerful AI capabilities to desktop environments. At its core is the NVIDIA GB10 Grace Blackwell Superchip, featuring a Blackwell GPU with fifth-generation Tensor Cores and FP4 precision support. This combination delivers up to 1,000 TOPS (trillion operations per second) of AI compute, making it well-suited for fine-tuning and inference tasks involving modern AI models such as NVIDIA's Cosmos Reason and GR00T N1 foundation models. A key feature is the GB10 Superchip's use of NVIDIA NVLink-C2C interconnect technology, which provides a CPU-GPU coherent memory model with five times the bandwidth of PCIe 5.0, significantly improving memory access speeds for AI workloads. DGX Spark users can also transition seamlessly to cloud-based solutions, with NVIDIA's AI software stack allowing models developed on DGX Spark to be deployed on DGX Cloud or other accelerated computing environments with minimal modifications.
The DGX Station offers a more powerful alternative, built around the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip designed to bring data-center-level AI performance to the desktop. This system boasts 784 GB of unified memory, a crucial advantage for AI workloads requiring extensive datasets. The GB300 Superchip integrates an NVIDIA Blackwell Ultra GPU with the latest Tensor Core technology and FP4 precision, linked to an NVIDIA Grace CPU using NVLink-C2C for fast and efficient data exchange. Equipped with the NVIDIA ConnectX-8 SuperNIC, the DGX Station supports speeds of up to 800Gb/s, enabling high-speed connectivity between multiple systems and allowing users to scale their AI workloads by interconnecting multiple machines to create an in-house AI cluster. The system is backed by NVIDIA's CUDA-X AI platform and supports NVIDIA NIM microservices via the NVIDIA AI Enterprise platform, providing pre-optimized inference microservices that streamline deployment workflows.
NVIDIA has collaborated with major system builders including ASUS, Dell, HP, and Lenovo to manufacture DGX Spark and DGX Station units. Reservations for DGX Spark are open immediately, while DGX Station is expected to be available later this year from partners such as ASUS, BOXX, Dell, HP, Lambda, and Supermicro. As AI workloads continue to grow in complexity, these systems provide a more accessible alternative to cloud-based AI solutions, allowing researchers and developers to work with advanced models without relying solely on remote infrastructure.