Researchers worldwide are utilizing accelerated computing to push the boundaries of science in quantum physics, digital biology, and climate research. NVIDIA announced at SC25 conference over 80 new scientific systems powered by their platform, totaling 4,500 exaflops of AI performance. The Horizon system at TACC will be America’s largest academic supercomputer when it comes online in 2026, supporting breakthroughs in various scientific fields.

The U.S. DOE partnered with NVIDIA to build seven new AI supercomputers at Argonne National Lab and Los Alamos National Lab. Systems like Solstice at ANL will feature 100,000 NVIDIA Blackwell GPUs, reaching 1,000 exaflops of AI training compute. LANL’s Mission and Vision systems will support classified applications and open science research, expected to be operational in 2027.

In Europe, the Jülich Supercomputing Centre’s JUPITER system achieved exaflop performance, Europe’s first exascale computer. JUPITER is used for global climate simulation and enables European researchers to run AI models and simulations at new levels of complexity. Other supercomputers like Blue Lion, Gefion, and Isambard-AI are also driving scientific research across the continent.

In Japan, RIKEN announced new supercomputers for AI for science and quantum computing, while Tokyo University of Technology built an AI supercomputer with NVIDIA DGX B200 systems. South Korea plans to deploy over 50,000 NVIDIA GPUs across sovereign clouds and AI factories, while Taiwan partners with Foxconn Hon Hai Technology group to build an AI factory supercomputer with 10,000 NVIDIA Blackwell GPUs for innovation.

Read more at NVIDIA: NVIDIA Accelerates AI for Over 80 New Science Systems Worldwide