NVIDIA has begun shipping DGX Spark, the world's smallest AI supercomputer, delivering petaflop-level performance in a compact desktop form factor. Built on the Grace Blackwell architecture, this system integrates NVIDIA GPUs, CPUs, networking, CUDA libraries, and AI software, enabling developers to run inference on models up to 200 billion parameters and fine-tune those up to 70 billion locally, accelerating agentic and physical AI development.
AI workloads are rapidly exceeding the capabilities of traditional PCs and laptops, pushing developers toward cloud or data center resources. DGX Spark addresses this by providing data-center-class power in a desktop-sized system, allowing teams to prototype, fine-tune, and deploy AI models locally. This evolution from the 2016 DGX-1—whose first unit sparked ChatGPT at OpenAI—returns to NVIDIA's mission of equipping every developer with supercomputer access to drive AI breakthroughs.
DGX Spark combines the complete NVIDIA AI ecosystem, including the GB10 Grace Blackwell Superchip, ConnectX-7 200 Gb/s networking, and NVLink-C2C for 5x PCIe Gen5 bandwidth. With 128GB coherent memory, it handles complex tasks like customizing FLUX.1 models for image generation or building vision agents using NVIDIA Cosmos Reason. Preloaded DGX OS (Ubuntu-based) and tools like NIM microservices enable out-of-the-box AI development for agentic applications.
Available from NVIDIA.com and partners like Micro Center in the U.S., DGX Spark systems from Acer, ASUS, Dell Technologies, GIGABYTE, HP, Lenovo, and MSI expand accessibility. Early adopters including Anaconda, Cadence, Google, Hugging Face, Meta, Microsoft, and research labs like NYU Global Frontier are optimizing tools and models. This rollout ignites innovation by bringing peta-scale computing to desktops for privacy-sensitive fields like healthcare.
NVIDIA's DGX Spark marks a pivotal step in personal AI computing, empowering developers worldwide to innovate without infrastructure barriers. By scaling the Grace Blackwell platform to desktops, it fosters rapid experimentation and deployment, positioning AI development at the fingertips of creators and researchers.
NVIDIA is the world leader in AI and accelerated computing.
TechIntelPro empowers B2B technology decision-makers with the latest industry trends, deep insights, leadership perspectives and actionable intelligence to drive both organizational and professional milestones—and thrive on the tech pulse.