Summary of the news:
NVIDIA today announced the launch of NVIDIA DGX Spark™, the world’s smallest AI supercomputer.
AI workloads are quickly outpacing the memory and software capabilities of the PCs, workstations, and laptops that millions of developers rely on today, forcing teams to move their work to the cloud or local data centers.
As a new class of computer, DGX Spark delivers a petaflop of AI performance and 128 GB of unified memory in a compact desktop form factor, giving developers the power to run inference on AI models with up to 200 billion parameters and locally refine models with up to 70 billion parameters. Additionally, DGX Spark allows developers to create AI agents and run advanced software stacks locally.
“In 2016, we built DGX-1 to give AI researchers their own supercomputer. I delivered the first system to Elon at a small startup called OpenAI – and from there ChatGPT was born, which started the AI revolution,” said Jensen Huang, founder and CEO of NVIDIA. “DGX-1 launched the era of AI supercomputers and unlocked the laws of scaling that govern modern AI. With DGX Spark, we return to that mission: putting an AI computer in the hands of every developer to spark the next wave of breakthroughs.”
DGX Spark brings together the entire NVIDIA AI platform, including GPUs, CPUs, networking and CUDA.® libraries and the NVIDIA AI software stack — in a system small enough for a lab or office, but powerful enough to accelerate the development of agentic and physical AI. By combining breakthrough performance with the reach of the NVIDIA ecosystem, DGX Spark transforms the desktop into an AI development platform.
DGX Spark systems deliver up to 1 petaflop of AI performance, accelerated by an NVIDIA GB10 Grace Blackwell superchip, NVIDIA ConnectX®-7 200 Gbps networking and NVIDIA NVLink™-C2C technology, delivering 5x the bandwidth of fifth-generation PCIe with 128GB of CPU-GPU coherent memory.
The NVIDIA AI software stack comes pre-installed to allow developers to start working on AI projects from the start. With DGX Spark, developers can access tools from the NVIDIA AI ecosystem, including NVIDIA NIM™ models, libraries, and microservices, enabling local workflows such as customizing Black Forest Labs’ FLUX.1 models to fine-tune image generation, creating a vision search and synthesis agent using the NVIDIA Cosmos™ vision language model Reason, or creating an AI chatbot using Qwen3 optimized for DGX Spark.
To celebrate shipping the DGX Spark around the world, Huang hand-delivered one of the first DGX Spark units to Elon Musk, Chief Engineer at SpaceX, today at Starbase, Texas. This exchange was linked to the origins of the supercomputer, as Musk was part of the team that received the first NVIDIA DGX™-1 supercomputer from Huang in 2016.
Other early DGX Spark beneficiaries include Anaconda, Cadence, Comfortable user interface, DockerGoogle, Hugging Face, JetBrains, LM Studio, Meta, Microsoft, Ollama and Robofluxtest, validate and optimize their tools, software and models for DGX Spark.
Research organizations around the world, including the NYU Global Frontier Lab, have previewed DGX Spark to boost their AI development.
“DGX Spark allows us to access peta-scale computing on our desktop,” said Kyunghyun Cho, professor of computer science and data science at NYU Global Frontier Lab. “This new way of conducting AI research and development allows us to rapidly prototype and experiment with advanced AI algorithms and models, even for privacy- and security-sensitive applications, such as healthcare. »
NVIDIA DGX Spark is now available
From Wednesday October 15, DGX Spark can be ordered on NVIDIA.com. Partner systems will be available from Acer, ASUS, Dell Technologies, GIGABYTE, HP, Lenovo, MSI, Micro Center stores in the United States and NVIDIA channel partners worldwide.
Pokémon TCG Pocket Announces “New and Updated” Features for Its First Anniversary Nintendo LifeShaymin and Absol star in this Pokémon…
Hundreds of satellites could soon be flying in orbital regions already too crowded to allow safe, long-term operations, a new…
The entrance to NASA's Jet Propulsion Laboratory in Pasadena, California, February 7, 2024. Robyn Beck | AFP | Getty ImagesNASA's…
The average price of new cars in the United States crossed the $50,000 mark for the first time in September,…
The Seattle Mariners are certainly showing no signs of fatigue after a 15-inning marathon and another critical playoff game less…
UPDATED with latest: Even before the first drop of rain falls, the impacts of the coming storm are already beginning…