Sparxity (formerly SpiNNcloud), a leader in sparse-native computing, today announced the deployment of its sparse-native computing rack system at the Jülich Supercomputing Centre (JSC), one of Europe’s largest supercomputing research centers and an institute of Forschungszentrum Jülich. The multi-million-dollar installation at JSC is built on Sparxity’s sparse-native systems for HPC and AI and will be utilized for running sparse machine learning models and large-scale HPC simulations. This implementation validates Sparxity’s vision to only compute what matters and coincides with the launch of its new brand identity, opening a new world of compute built on sparsity.
This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20261028075211/en/
JSC is the home of Europe’s first exascale supercomputer and is a leader in AI with the JUPITER AI Factory, one of 19 AI factories in Europe and the only one built around an exascale machine. The new Sparxity system at JSC is event-based and computes only when something happens, a form of sparsity in practice. The system installed at JSC, named JUBIS, is designed to deliver 10X faster sparse solver execution, 18X lower energy consumption for event-based ML workloads and 20X shorter time to an acceptable solution in optimization problems. The system is intended to become available for wider HPC and AI applications and the installation at JSC is another proof point that sparse-native computing works in real infrastructure.
“Sparse machine learning represents one of the biggest opportunities to fundamentally improve the efficiency of AI,” said Dr. Guido Trensch, Simulation and Data Lab Neuroscience, Jülich Supercomputing Centre. “By adopting Sparxity, we’re able to complement our globally unique HPC infrastructure with a fundamentally new approach, opening up new research opportunities so that we can further develop advancements in neuroscience.”
As AI infrastructure faces challenges around energy consumption and cost, sparsity is one of the biggest untapped routes for AI infrastructure. Research has shown that AI models can be sparser than they are today, but hardware built for dense computation has held them back. Sparse computation takes an approach of reducing the amount of computing needed and the amount of data that needs to be moved to reduce bottlenecks.
“Model builders have spent years designing around the limits of dense hardware,” said Sparxity co-founder and CEO Hector Gonzalez. “Sparxity exists to unlock what they would build without those limits, by computing only what matters. Our rebrand focuses entirely on the sparse-native approach as a practical foundation for the next generation of AI infrastructure. Our current systems bring sparsity to both HPC and AI, and our collaboration with Jülich Supercomputing Centre, one of the largest supercomputing facilities in Europe, shows the value of that approach at supercomputer scale.”
Sparxity builds full-stack systems around this principle, from silicon to software, rather than a chip alone. Its current systems serve HPC and AI, and its next generation commits fully to sparsity for AI inference. For model builders, that means freedom from the constraints of dense hardware and access to finer, more dynamic sparsity. For AI inference operators, such as datacenters and neoclouds, sparsity means more tokens at higher speed and lower cost, without trade-offs.
To learn more about Sparxity and its sparse-native computing systems, please visit sparxity.ai.
The Sparxity system at JSC is supported by the European Regional Development Fund (EFRE) and the State of North Rhine-Westphalia (NRW) under the ‘EFRE/JTF-Programm NRW 2021–2027’ program.
About Sparxity
Sparxity builds sparse-native compute for AI. Our principle is simple: compute only what matters. For years, models have been shaped by the dense hardware they run on. Sparxity builds the hardware that lets them go further: full-stack systems, from custom chips to software, built for fine-grained, dynamic sparsity, where intelligence scales without trading off speed or efficiency. Deployed in rack- and supercomputer-scale systems across Europe and the US, Sparxity opens a new world of compute for the builders shaping what comes next. Learn more at sparxity.ai.
View source version on businesswire.com: https://www.businesswire.com/news/home/20261028075211/en/
Media gallery
