Ten years ago at NVIDIA GTC 2016, Jensen Huang introduced DGX-1, our first AI supercomputer. Since then, NVIDIA DGX has transformed from a single system into a blueprint for AI factories. What hasn’t changed is the reason why we build DGX: to test the systems and work through the details of interconnects, cooling, and software so our partners and customers have a proven foundation to build on. Charlie Boyle, our VP of DGX Systems, looks back at how DGX got its start, what’s changed, and what comes next. Thanks to everyone who’s been on this journey with us. #DecadeOfDGX
Congratulations to NVIDIA on continuing to redefine what is possible with AI and accelerated computing. At RIVSAL INDUSTRIES, we would like to invite NVIDIA to learn about a disruptive technology that could help address one of the greatest challenges facing the next generation of AI infrastructure: continuous energy availability directly at the point of consumption. Our E-500 provides 500 kVA of nominal capacity per module, operates 24/7/365, uses approximately 20 liters of water per day, requires no fossil fuels and produces zero direct CO₂ emissions during operation. Its modular architecture allows multiple units to be integrated according to the power requirements of each project, opening the possibility of supporting large-scale AI data centers. We are not seeking to replace solar, BESS or existing energy infrastructure. Our vision is to complement them with continuous on-site generation, helping create cleaner, more resilient and scalable energy architectures. AI is transforming the world. The energy infrastructure supporting it must evolve as well. We would be honored to present our technology to NVIDIA and explore its potential applications together. #NVIDIA #DataCenters #CleanEnergy #EnergyInnovation #Sustainability
What is truly impressive is that the DGX has evolved from an AI supercomputer into a vital component of modern AI infrastructure. I look forward to the technological evolution of the next decade.
Ten years from a single AI supercomputer to the blueprint for 'AI factories' is a remarkable trajectory. As someone building in tech, it's a good reminder that foundational infrastructure bets — even ones that seem niche at the time — can define an entire industry's next decade. Congrats to the DGX team
DGX shows why AI infrastructure is rarely just a compute problem. Interconnects, cooling and software determine how much of that compute can actually be used in production. Proving those dependencies together before deployment can remove some of the most expensive infrastructure surprises later.
It’s easy to focus on the GPUs when looking back at DGX, but the interconnects, cooling, and software are probably where a lot of the hard engineering work actually happened. Ten years of figuring out those details is a pretty good reminder that scaling AI has never been just a compute problem.
It’s impressive to see how DGX evolved from a single AI system into a foundation for AI infrastructure. The next phase of AI growth will likely depend not only on models, but also on the computing platforms that power them.
A remarkable decade of innovation. DGX has played a key role in advancing AI infrastructure, showing that breakthroughs require not only powerful hardware, but also the engineering foundation behind scalable AI systems
From DGX-1 in 2016 to blueprint for AI factories today — a decade that defined the future. What hit me: "proven foundation to build on" — testing interconnects, cooling, software so partners can scale. That's real infrastructure thinking. Exactly our philosophy at FinTech Platform. I'm Ali Ikram, Founder — building an AI-Native Global Multi-Asset Hybrid Exchange, more than an exchange, a Global Intelligence Layer. Like NVIDIA built foundation for AI, we build foundation for Finance: Intelligence Layer: Data → Risk Visibility → Better Decisions Infrastructure Layer: Access, Liquidity, Execution, Clearing, Settlement, Custody, Payments, Risk & Compliance DGX = Blueprint for AI Factories. Next 10 years = Blueprint for Intelligent Financial Factories. Congrats to Jensen Huang, Charlie Boyle & NVIDIA team on #DecadeOfDGX Ali Ikram | Founder, FinTech Platform — Building Intelligence Layer for Global Finance
The evolution from one system to an entire blueprint for AI infrastructure is pretty impressive. The focus on solving the hard details before scaling them is what really stands out. 👏
Thank you to our customers including BMW Group, BNY, DeepL, Eli Lilly and Company, Foxconn, Linköping University, Meta, MITRE, OpenAI and University of Florida, who use #NVIDIADGX to solve their hardest challenges.