Nvidia has announced the global availability of DGX Spark, describing it as the world’s smallest Ai supercomputer. The system delivers a petaflop of performance in a compact desktop form factor, aimed at developers building and fine-tuning large-scale models and agents.
The company said DGX Spark is designed to address the growing gap between the needs of Ai workloads and the capacity of traditional PCs, workstations and laptops.
A New Class Of Computer
DGX Spark integrates Nvidia GPUs, CPUs, networking, CUDA libraries and the company’s Ai software stack. The system provides 128GB of unified memory and can run inference on models with up to 200 billion parameters, while fine-tuning models of up to 70 billion parameters locally.
Nvidia said this enables developers to run agentic and physical Ai applications without relying on cloud or datacentre infrastructure.
“In 2016, we built DGX-1 to give Ai researchers their own supercomputer. I hand-delivered the first system to Elon at a small startup called OpenAI — and from it came ChatGPT,” said Jensen Huang, Founder and CEO of Nvidia. “With DGX Spark, we return to that mission — placing an Ai computer in the hands of every developer to ignite the next wave of breakthroughs.”
Technical Specifications
The DGX Spark is powered by the GB10 Grace Blackwell Superchip, paired with ConnectX-7 200 Gb/s networking and NVLink-C2C technology. Nvidia said this delivers five times the bandwidth of PCIe Gen 5, with CPU-GPU coherent memory across 128GB.
The Ai software stack comes preinstalled, giving developers immediate access to models, libraries and Nvidia NIM microservices. Use cases include image generation using Black Forest Labs’ FLUX.1 models, building vision search agents with the Cosmos Reason model, or deploying chatbots based on Qwen3.
Early Access And Ecosystem
To mark the launch, Huang hand-delivered one of the first DGX Spark units to Elon Musk at SpaceX in Texas, echoing the original delivery of DGX-1 in 2016.
Nvidia said early adopters include Anaconda, Cadence, ComfyUI, Docker, Google, Hugging Face, JetBrains, LM Studio, Meta, Microsoft, Ollama and Roboflow. These organisations are testing and optimising tools and models for the new system.
Research Applications
The NYU Global Frontier Lab previewed DGX Spark as part of its Ai research program. Kyunghyun Cho, Professor of Computer and Data Science at NYU, said the system provided new opportunities for experimentation.
Related: Best Business Laptops for Work & School
Related: Best Gaming Laptops
Related: Best Portable Laptop
“DGX Spark allows us to access peta-scale computing on our desktop,” Cho said. “This enables us to rapidly prototype and experiment with advanced Ai algorithms and models, even for privacy- and security-sensitive applications such as healthcare.”
Availability
DGX Spark will be available for order from 15 October on Nvidia.com. Systems built by partners including Acer, Asus, Dell Technologies, Gigabyte, HP, Lenovo and MSI will also be available through retail and channel partners worldwide, including Micro Center in the US.
Last Updated on 14 October 2025 by Nick Ross
