Nine months ago, our site in Ward County was an empty steel shell in West Texas. Today it’s a live AI data center delivering thousands of GPUs.
That’s the pace the AI build-out demands. And it’s the pace we’re delivering. In the last month alone, Nscale has deployed 24,000 GPUs. That nearly doubles our deployments this year to over 50,000 GPUs powering production workloads across six active data centers in five countries.
To put that GPU count in context, here’s why it matters: GPUs are the engines behind modern AI. They handle thousands of calculations at the same time, which is what it takes to train a model and serve it to millions of people. Every AI tool and service you use today runs on them. More GPUs online means more AI people can actually use.
Scaling to more than 50,000 GPUs in less than nine months demonstrates our ability to execute. Our speed shows that we are able to stand up infrastructure at a rate the market rarely sees.
And our deep alignment with the world’s leading hyperscalers and frontier AI labs shows conviction in Nscale’s platform. With over $103 billion in TCV, our long-term partnerships will enable us to build the foundation for a broader enterprise market.
None of this happens without the right people. Nidhi Chappell, our President of AI Infrastructure, helped grow Microsoft’s Azure AI capacity from a few thousand GPUs to the world’s largest fleet. Sam Huckaby, our President of Data Centers, led global data center development at Oracle after delivering hyperscale campuses at Vantage. They have both done this at scale before, and it shows.
This is the fourth industrial revolution. We have doubled our fleet in a short amount of time and we are just getting started.