Elon Musk Will get Simply-Launched NVIDIA DGX Spark: Petaflop AI Supercomputer Lands at SpaceX

The following AI revolution begins the place rockets launch. NVIDIA DGX Spark’s first cease: Starbase, Texas.

NVIDIA founder and CEO Jensen Huang arrived on the SpaceX facility — amid towering engines and gleaming metal — to hand-deliver the corporate’s just-launched DGX Spark to Elon Musk.

Huang arrived strolling previous rows of engineers who waved and grinned. Moments later, Musk appeared within the cafeteria, greeting workers and opening donuts and chips for youths earlier than grabbing a slice of pizza.

Huang joined him, recounting the story of delivering the primary DGX system to OpenAI and explaining how Spark takes that mission additional.

“Think about delivering the smallest supercomputer subsequent to the largest rocket,” Huang mentioned with amusing.

The handoff got here as SpaceX ready for the eleventh take a look at of Starship, the world’s strongest launch car.

DGX Spark packs 128GB of unified reminiscence and delivers a petaflop of AI efficiency, sufficient to run fashions with 200 billion parameters domestically.

Constructed for builders, researchers and creators, DGX Spark brings supercomputer-class efficiency past the information middle — able to seize and go.

9 years in the past, NVIDIA guess on the way forward for AI with NVIDIA DGX-1. Right now, that guess goes past the information middle with the handoff to Musk coming amid the eleventh take a look at of SpaceX’s Starship, the world’s strongest launch car.

From robotics labs to inventive studios, DGX Sparks are touchdown the place concepts occur… placing petaflop AI inside arm’s attain of everybody.

This weblog might be up to date as DGX Spark programs land from Ollama in Palo Alto to Arizona State’s robotics lab, from Refik Anadol’s studio to the fingers of Jo Mardall at Zipline. Every supply is a brand new chapter within the story of AI.

What’s Inside DGX Spark? A Supercomputer within the Palm of Your Hand

It’s the scale of a chunk of origami paper, and the thickness of a hardcover ebook. It acts like a rocket engine for AI.

DGX Spark isn’t simply compact — it’s dense with chance. Inside its 1.2 kg chassis is a full-blown AI supercomputer constructed to take AI past the information middle and into the fingers of those that create.

At its core:

• NVIDIA GB10 Grace Blackwell Superchip — delivering as much as 1 petaflop of AI efficiency at FP4 precision.
• 128GB of unified CPU-GPU reminiscence — so builders can prototype, fine-tune and run inference domestically with out bouncing between machines or cloud cases.
• NVIDIA ConnectX networking for clustering and NVIDIA NVLink-C2C for 5x PCIe bandwidth.
• NVMe storage for pace and HDMI out for visuals.

And it’s not simply {hardware}. DGX Spark comes with the complete NVIDIA AI software program stack — frameworks, libraries, pretrained fashions and NVIDIA NIM microservices, able to energy workflows like:

• Customizing image-generation fashions resembling FLUX.1
• Constructing imaginative and prescient search and summarization brokers with NVIDIA Cosmos
• Deploying optimized chatbots utilizing Qwen3

This isn’t a dev field. It’s a launchpad… A petaflop of AI efficiency inside arm’s attain for builders, researchers and creators in all places.

Companions Energy Up: DGX Spark Lands Past the Information Middle

From PC giants to AI pioneers, the diminutive DGX Spark is certain to start out one thing large.

DGX Spark is already within the fingers of innovators — from ISVs optimizing their instruments to researchers pushing the boundaries of robotics, artwork and edge AI.

DGX Spark isn’t only a breakthrough in dimension and efficiency… it’s a platform constructed on collaboration. Acer, ASUS, Dell Applied sciences, GIGABYTE, HP, Lenovo and MSI are rolling out programs that put petaflop AI in your desk, reworking the desktop into an AI launchpad.

These companions ship greater than {hardware}. They ship chance — NVIDIA’s full AI stack in a compact kind issue that accelerates agentic and bodily AI improvement in all places concepts occur.

And the rollout doesn’t cease at OEMs. DGX Sparks are already lighting up the AI ecosystem. Some highlights:

• To be in Palo Alto, rewriting how builders run giant language fashions domestically.
• NYU International Frontier Labthe place researchers prototype algorithms for privacy-sensitive functions.
• Ziplinepushing the boundaries of autonomous supply.
• Arizona State Collegeworking robotics simulations and imaginative and prescient fashions on the edge.
• Refik Anadol’s studiomixing artwork and AI with petaflop efficiency.

From inventive studios to robotics labs, DGX Sparks are touchdown the place creativeness meets engineering — taking AI past the information middle and into the fingers of these constructing what’s subsequent.

DGX Spark Ignites a New Period of Desktop AI

As DGX Spark programs land throughout the ecosystem, six early companions are already displaying what’s attainable when petaflop-class AI meets the desktop:

• Dell Professional Max with GB10 powered by Grace Blackwell and DGX OS brings data-center class AI at your desk, delivering as much as 200B parameters- with unified reminiscence, CUDA and on the spot scalability. For these needing much more energy, connecting two Dell Professional Max with GB10 programs turns them right into a single node able to supporting 400 billion-parameter fashions.
• HP’s ZGX Nano G1n AI Station brings severe AI horsepower to the desktop — petaflop-class efficiency, 128GB unified reminiscence, and full entry to NVIDIA’s AI stack.
• Lenovo ThinkStation PGX brings agentic AI to the desktop. Constructed for researchers and builders, it runs fashions as much as 200B parameters and seamlessly bridges native prototyping with cloud deployment.
• LM Studio is now reside on DGX Spark, transport for Linux on ARM. Builders can spin up non-public LLM servers, run fashions like Qwen3 Coder domestically, and join throughout the community—no cloud required.
• Python on DGX Spark is already altering the sport. Anaconda’s Stanley Seibert dives into unified reminiscence, hybrid compute, and what it means for working billion-row dataframes and billion-parameter fashions — all domestically.
• Roboflow brings imaginative and prescient AI to the desktop. With DGX Spark, builders can now practice and fine-tune extremely correct pc imaginative and prescient fashions domestically, giving their software program the sense of sight and powering clever, linked programs.
• Additionally notable: Serve Robotics accomplished the first-ever DGX Spark deliveries: The corporate’s autonomous sidewalk robots, powered by NVIDIA Jetson Orin and skilled in Isaac Sim, carried out the primary deliveries of DGX Spark programs to builders — a milestone that actually brings the following era of AI to their doorsteps.

From Palo Alto to the PyData ecosystem, DGX Spark is redefining what’s attainable by placing supercomputing energy inside arm’s attain of each creator, researcher and builder.

‘Freaking Cool’: DGX Spark Early Opinions Are In

As DGX Spark programs start touchdown, early reviewers are weighing in — and the decision is obvious: this tiny AI supercomputer is making an enormous influence. From developer-first design to petaflop efficiency in a desktop kind issue, Spark, is incomes excessive marks throughout the board. A couple of fast hits:

HotHardware Really useful

• NVIDIA’s Construct portal playbooks are clear, thorough, and beginner-friendly.
• “Even novices can have no hassle getting began.” — HotHardware
• “Goal-built {hardware} like DGX Spark will grow to be the norm.”
• “That is the place to start out.”

ServeTheHome

• “Freaking cool.”
• 128GB unified reminiscence + NVIDIA Blackwell = no want for cloud or further accelerators.
• “Should-have for AI devs — and execs bringing AI into their orgs.”

Level1Techs

• “Soup-to-nuts dev lab.”
• Construct RAG, multimodal brokers, video search — day one.
• “A lab in a field.”

Boyan Tunguz

• “Each native DS/ML/AI dev’s dream.”
• “{Hardware} + software program = magic.”

NetworkChuck

• “For those who’re fine-tuning and doing enjoyable knowledge science stuff each day — that is your machine.”

DGX Spark might be typically obtainable beginning Wednesday, Oct. 15, on NVIDIA.com and thru companions worldwide.

Delivering NVIDIA DGX Spark to AI Leaders

From a music studio in Los Angeles to a robotics lab in Arizona, a brand new form of machine is touchdown on desks in all places. NVIDIA’s DGX Spark provides creators, researchers and tinkerers a petaflop of AI energy sufficiently small to hold underneath one arm. On this brief movie, will.i.am, Zipline, and Arizona State College present how they’re already placing it to work — turning concepts as soon as reserved for knowledge facilities into experiments anybody can run.