Information Abstract:
- NVIDIA founder and CEO Jensen Huang delivers DGX Spark to Elon Musk at SpaceX.
- This week, NVIDIA and its companions are delivery DGX Spark, the world’s smallest AI supercomputer, delivering NVIDIA’s AI stack in a compact desktop type issue.
- Acer, ASUS, Dell Applied sciences, GIGABYTE, HPI, Lenovo and MSI debut DGX Spark methods, increasing entry to highly effective AI computing.
- Constructed on the NVIDIA Grace Blackwell structure, DGX Spark integrates NVIDIA GPUs, CPUs, networking, CUDA libraries and NVIDIA AI software program, accelerating agentic and bodily AI growth.
NVIDIA right now introduced it’s going to begin delivery NVIDIA DGX Spark™, the world’s smallest AI supercomputer.
AI workloads are shortly outgrowing the reminiscence and software program capabilities of the PCs, workstations and laptops hundreds of thousands of builders depend on right now — forcing groups to shift work to the cloud or native knowledge facilities.
As a brand new class of laptop, DGX Spark delivers a petaflop of AI efficiency and 128GB of unified reminiscence in a compact desktop type issue, giving builders the ability to run inference on AI fashions with as much as 200 billion parameters and fine-tune fashions of as much as 70 billion parameters domestically. As well as, DGX Spark lets builders create AI brokers and run superior software program stacks domestically.
“In 2016, we constructed DGX-1 to offer AI researchers their very own supercomputer. I hand-delivered the primary system to Elon at a small startup referred to as OpenAI — and from it got here ChatGPT, kickstarting the AI revolution,” mentioned Jensen Huang, founder and CEO of NVIDIA. “DGX-1 launched the period of AI supercomputers and unlocked the scaling legal guidelines that drive trendy AI. With DGX Spark, we return to that mission — putting an AI laptop within the fingers of each developer to ignite the subsequent wave of breakthroughs.”
DGX Spark brings collectively the total NVIDIA AI platform — together with GPUs, CPUs, networking, CUDA® libraries and the NVIDIA AI software program stack — right into a system sufficiently small for a lab or an workplace, but highly effective sufficient to speed up agentic and bodily AI growth. By combining breakthrough efficiency with the attain of the NVIDIA ecosystem, DGX Spark transforms the desktop into an AI growth platform.
DGX Spark methods ship as much as 1 petaflop of AI efficiency, accelerated by a NVIDIA GB10 Grace Blackwell Superchip, NVIDIA ConnectX®-7 200 Gb/s networking and NVIDIA NVLink™-C2C expertise, offering 5x the bandwidth of fifth-generation PCIe with 128GB of CPU-GPU coherent reminiscence.
The NVIDIA AI software program stack is preinstalled to allow builders to start out engaged on AI initiatives out of the field. With DGX Spark, builders can entry NVIDIA AI ecosystem instruments together with fashions, libraries, and NVIDIA NIM™ microservices, enabling native workflows similar to customizing Black Forest Labs’ FLUX.1 fashions to refine picture era, making a imaginative and prescient search and summarization agent utilizing the NVIDIA Cosmos™ Purpose imaginative and prescient language mannequin, or constructing an AI chatbot utilizing Qwen3 that’s optimized for DGX Spark.
To rejoice DGX Spark delivery worldwide, Huang hand-delivered one of many first items of DGX Spark to Elon Musk, chief engineer at SpaceX, right now in Starbase, Texas. The change was a connection to the supercomputer’s origins, as Musk was among the many group that obtained the primary NVIDIA DGX™-1 supercomputer from Huang in 2016.
Different early recipients of DGX Spark, together with Anaconda, Cadence, ComfyUI, Docker, Google, Hugging Face, JetBrains, LM Studio, Meta, Microsoft, Ollama and Roboflow, are testing, validating and optimizing their instruments, software program and fashions for DGX Spark.
Analysis organizations all over the world, together with the NYU International Frontier Lab, previewed DGX Spark to spice up their AI growth.
“DGX Spark permits us to entry peta-scale computing on our desktop,” mentioned Kyunghyun Cho, professor of laptop and knowledge science on the NYU International Frontier Lab. “This new strategy to conduct AI analysis and growth allows us to quickly prototype and experiment with superior AI algorithms and fashions — even for privacy- and security-sensitive purposes, similar to healthcare.”
NVIDIA DGX Spark Now Transport
Beginning Wednesday, Oct. 15, DGX Spark might be ordered on NVIDIA.com. Associate methods shall be accessible from Acer, ASUS, Dell Applied sciences, GIGABYTE, HP, Lenovo, MSI in addition to Micro Heart shops within the U.S., and from NVIDIA channel companions worldwide.
About The Author
