Europe is building a record 35 new NVIDIA AI supercomputers across 23 countries, delivering over 800 exaflops of AI performance and providing access to next-generation infrastructure for more than 3 million researchers. The systems, announced at ISC High Performance 2026, represent the continent's largest single-year expansion of supercomputing capacity and span national AI factories, supercomputing centers, and universities.
"AI is the new instrument of science, and Europe is building the infrastructure to put it in the hands of millions of researchers," said Jensen Huang, founder and CEO of NVIDIA. "With NVIDIA accelerated computing, researchers can simulate more complex systems, train scientific AI models and build agentic AI workflows that turn Europe's data and expertise into breakthroughs for the world."
The deployments rely on NVIDIA Blackwell and Hopper platforms, Quantum InfiniBand networking, CUDA-X libraries, NIM microservices, and AI Enterprise software. The full-stack approach covers model training, simulation, inference, and agentic AI. Among the flagship projects are Barcelona Supercomputing Center's MareNostrum5 AI upgrade, BavariaAI's Blue Swan, IT4LIA, HLRS's HammerHAI, and NAISS's Mimer EuroHPC AI Factory.
"BSC is committed to building AI infrastructure that advances science, industry and society," said Mateo Valero Cortés, director of the Barcelona Supercomputing Center. "With the upgrade to MareNostrum5 and NVIDIA accelerated computing, the consortium composed of Spain, Portugal and Türkiye will make available to European researchers the tools to tackle some of the world's most complex challenges, from climate modeling to biomedical discovery."
AI for climate and decarbonization
Siemens Energy is using the Siemens Xcelerator portfolio, accelerated by NVIDIA technologies including Omniverse libraries and CUDA-X, to unify design, simulation, and manufacturing for hydrogen-capable gas turbines. The workflow supports rapid, simulation-driven design iterations and cuts simulation times by as much as 77%, accelerating development of low-carbon gas turbines that can run on up to 100% hydrogen.
NVIDIA also supports initiatives applying AI to climate and Earth systems modeling, biomedical research, and clean-energy technologies such as fusion, hydrogen, and carbon capture. The scale of new supercomputing capacity makes it possible to run previously impractical simulations at high resolution.
Quantum-GPU supercomputing advances
European centers are integrating quantum processors with classical supercomputers using the CUDA-Q platform. CINECA and Pasqal are deploying a neutral-atom QPU, with CUDA-Q integrated through Slurm, for optimization and materials science workloads. Fraunhofer FOKUS is connecting CUDA-Q with the quantum programming language Eclipse Qrisp, making it easier for researchers to write and simulate complex quantum algorithms.
Barcelona Supercomputing Center deployed an analog quantum computer from Qilimanjaro Quantum Tech, which integrated CUDA-Q into its software development kit. Separately, researchers at Jülich Supercomputing Centre, working with NVIDIA, set a record by fully simulating a universal 50-qubit quantum computer on the JUPITER system powered by NVIDIA GH200 Grace Hopper Superchips.
Why this matters for IT and development teams
The rapid buildout of AI factories means IT and DevOps teams will need to manage, network, and secure massive GPU clusters running distributed training and inference workloads. For IT teams, the rollout demands deeper expertise in managing large-scale GPU clusters, AI factory operations, and hybrid quantum workflows-skill areas gaining focus in AI for IT & Development programs. Developers working on simulation, HPC, or quantum-accelerated applications will increasingly target CUDA-Q and full-stack NVIDIA environments that span on-premises and cloud-like AI services.
Your membership also unlocks:
