Juvoly, a Dutch AI startup, announced that on March 10, one of the first NVIDIA DGX B200 systems in the Netherlands was unveiled at NorthC Datacenters in Rotterdam by Prince Constantijn van Oranje — a special envoy for the Dutch deep tech startup ecosystem. These powerful systems are part of AI startup Juvoly’s new NVIDIA accelerated computing cluster, providing the computing power to develop the next generation of AI models locally and host them within the country. This marks a significant step toward an independent Dutch AI sector.
Juvoly, a Dutch startup and member of the NVIDIA Inception program, focuses on healthcare and will utilize these systems to train advanced speech recognition models.
“With this infrastructure, we can develop AI solutions that improve healthcare outcomes, without sensitive patient data and other information being sent abroad,” says Thomas Kluiters, co-founder of Juvoly.
The Importance of Data Sovereignty
European organizations are committed to maintaining data sovereignty, yet they often leverage American cloud providers for AI training and hosting.
To help keep data in Europe, supercomputers such as NVIDIA DGX B200 systems provide the computing power to develop and host large-scale AI models locally.
Investing in AI Infrastructure
According to Kluiters, the country has no shortage of technological or financial capabilities—only ambition: “If a startup less than two years old can bring the latest supercomputers to the Netherlands, then other companies can too.” He urges businesses and governments to invest in domestic AI infrastructure to strengthen the country’s technological sovereignty. By deploying its NVIDIA DGX AI supercomputer, Juvoly wishes to lead by example.
100% Dutch AI
In a short time, Juvoly has positioned itself as a pioneer in developing fully Dutch speech technology that contributes to a more inclusive society. The startup creates and hosts speech-to-text models optimized for language barriers, accents, and medical terminology. The result: a model that outperforms international players while consuming less energy—already processing over 5,000 general practitioner consultations per day.
Additionally, Juvoly is collaborating with leading Dutch universities Erasmus MC and TU Delft to develop a Medical Dutch Large Language Model (LLM), which is planned to be released as open-source in the future. With the deployment of its new supercomputers, Juvoly can develop this technology completely independently, without compromising on privacy or data security.