NSF LCCF Horizon Supercomputer To Power Breakthroughs for the Nation’s Leading Scientists

Installation of nation’s largest academic supercomputer begins, ushering in a new era of open science and discovery

    None
    Sabey Data Center, the new home of the NSF LCCF Horizon supercomputer. The campus is being built to accommodate high-density deployments, such as those required for AI applications with low-cost renewable power.

    Something big is coming for science.

    The U.S. National Science Foundation (NSF) launched the installation of Horizon, the nation’s largest academic supercomputer, at the Texas Advanced Computing Center (TACC) at The University of Texas at Austin. Horizon is the centerpiece of NSF’s new Leadership-Class Computing Facility (NSF LCCF), a national resource on a par with iconic scientific initiatives such as the James Webb Space Telescope and the IceCube Neutrino Observatory.

    When it enters production in Spring 2026, Horizon will provide U.S. researchers with unprecedented computing and artificial intelligence capabilities, enabling breakthroughs across physics, climate science, medicine, energy, and beyond.

    Dan Stanzione, Executive Director of the Texas Advanced Computing Center (left), and Michael Kratsios, Director of the White House Office of Science and Technology Policy (right), at TACC in Spring 2025.

    “Our partnership with Dell, NVIDIA, and VAST Data gives Horizon groundbreaking capabilities particularly in the use of AI for scientific innovation,” said Dan Stanzione, executive director of TACC and associate vice president for Research at UT Austin.

    “Our previous large scale systems have enabled the work of more than 100,000 students and researchers, leading to countless discoveries ranging from subtle changes in building codes to increase resilience in hurricanes to Nobel prize-winning research that advances our understanding of the universe. Horizon represents the largest investment the NSF has made in computing infrastructure, so we expect even more ground-breaking achievements. After years of planning, we are thrilled to now be building out the system.”

    Built for the Future of Discovery

    Developed in collaboration with Dell Technologies, NVIDIA, VAST Data, Spectra Logic, Versity, and Sabey Data Centers, the Horizon supercomputer combines cutting-edge technologies with advanced infrastructure to redefine what is possible in scientific computing.

    • Performance: 300 petaflops, delivering a 10x improvement in simulation speed over Frontera, the current No. 1 academic supercomputer in the U.S.
    • AI Power: 20 exaflops for AI at bf16/fp16 / 80 Exaflops for AI at FP4, more than 100x improvement over today’s systems.
    • Scale: NVIDIA Grace Blackwell platform and NVIDIA Vera CPU servers featuring 1 million CPU cores and 4,000 GPUs.
    • Networking: Interconnected by the NVIDIA Quantum-2 InfiniBand networking platform with In-Network Computing.
    • Local All-Solid State Storage: 400PB delivering well more than 10TB/s of read/write bandwidth along with multi-tenancy and Quality-of-Service capabilities.
    • Efficiency: Up to 6x more energy efficient, powered by a new 15-20 MW data center with advanced liquid cooling in Round Rock, Texas.

    “Powered by NVIDIA’s accelerated computing platform, Horizon combines the massive parallelism of our GPUs with high-performance networking and energy-efficient architectures to deliver transformative scale for AI and simulation,” said Ian Buck, vice president of hyperscale and HPC. “The system will enable U.S. researchers to model the planet’s climate, advance biomedical discovery and unlock insights across physics, energy and beyond — pushing the boundaries of what’s possible in science through AI-driven computing.”

    “Over the years, Dell Technologies and TACC have empowered many brilliant minds to work towards solving some of humanity's greatest challenges,” said Arun Narayanan, senior vice president, Compute and Networking, Dell Technologies. “Horizon, powered by Dell PowerEdge servers, will provide the next generation of scientists and researchers with advanced compute to accelerate the pace of research and unlock new scientific breakthroughs.”

    “Horizon represents the next frontier of open science, a national instrument for discovery,” said Jeff Denworth, co-founder, VAST Data. “We are witnessing an unprecedented paradigm shift in supercomputing, and TACC is a lighthouse example of building powerful data-computing systems that make no compromise on data management capability or system availability. With the VAST AI OS, TACC is leading a new era of deep discovery on an architecture built for massive computational parallelism and AI-based data contextualization.”

    Deployed in September 2025, the NSF LCCF Ranch archive system delivers massive, high-performance storage and tape-based archiving for research data.

    Exabyte-Scale Storage: Ranch In Production

    Alongside Horizon, NSF and TACC have already deployed Ranch, the first production system of the LCCF and the largest academic data storage system in the nation. This first-of-its-kind endeavor for science that establishes a new industry standard for data management at an exabyte scale.

    With more than one exabyte of capacity, Ranch supports both new research and decades of scientific data, combining high-speed flash storage with sustainable tape-based archiving. Researchers can store, access, and preserve data securely for the long term, ensuring science is reproducible and future-proof.

    Built By Science For Science

    The NSF LCCF is designed from the ground up to serve researchers’ most challenging computational problems. At its heart is the Characteristic Science Applications (CSA) program, a set of 11 groundbreaking projects that continue to shape Horizon’s design and drive discoveries across health to astrophysics.

    “In addition to ensuring that we assess the capability of the LCCF in terms of real science, the team is leading community researchers through the process of gaining real-world experience tuning applications for the new architecture,” John West, NSF LCCF deputy director, explained. “Our experiences from these codes will then shape the training and resources we provide to users when the machine becomes available, helping to ensure that communities can get up to speed quickly on the new system.”

    Serving the Nation’s Research Community

    Cohort 3 of the U.S. National Science Foundation Leadership-Class Computing Facility Internship. Interns train on world-class computing systems, strengthen their workforce skills, and build lasting professional networks.

    The NSF LCCF is open to U.S. researchers, engineers, and international collaborators on NSF-funded projects. Supporting 10,000 to 15,000 users annually — from students and faculty members to national labs, nonprofits, and industry — the facility will allocate resources through a rigorous peer-review process, ensuring access for the most innovative and impactful research.

    The NSF Leadership-Class Computing Facility at TACC is supported by NSF Award #2323116.

    About the Texas Advanced Computing Center (TACC)
    The Texas Advanced Computing Center at The University of Texas at Austin is the nation’s leading academic supercomputing center. Since 2001, TACC has advanced discoveries across scientific disciplines by providing world-class systems, software, and expertise to researchers addressing society’s greatest challenges. TACC offers high performance computing, AI at scale, storage, visualization, training, and workforce development, fostering innovation that transforms science and improves lives. As home to the U.S. National Science Foundation Leadership-Class Computing Facility, TACC will drive the next decade of breakthroughs in computational research and discovery. For more information, visit www.tacc.utexas.edu.