Argonne Leadership Computing Facility

06/19/2026

06/19/2026

Join Argonne National Laboratory at ISC High Performance, June 22-26, in Hamburg, Germany.

Throughout the week, Argonne researchers, including several ALCF staff members, will present work spanning AI for science, agentic workflows, next-generation supercomputers, quantum computing, and emerging AI accelerators. We're looking forward to connecting with the global HPC community and sharing the lab's latest advances in scientific computing through papers, panels, tutorials, workshops, and Birds of a Feather sessions.

For a list of Argonne's conference activities, visit: https://www.alcf.anl.gov/argonne-isc-2026

For the full program, visit: https://isc-hpc.com/

06/18/2026

Using ALCF supercomputing resources, a team led by the University of Southern California developed Allegro-FM, an AI foundation model for exascale molecular dynamics simulations that can describe a wide range of materials properties and processes. Demonstrated on Argonne National Laboratory's Aurora system at scales of more than 4 billion atoms, Allegro-FM accurately captured structural, mechanical, and thermodynamic properties, even for tasks on which the model was not trained, highlighting the potential of foundation models to accelerate materials design and discovery.

https://ar25.alcf.anl.gov/science/highlights/nakano

06/17/2026

Today at HPE Discover 2026, Argonne National Laboratory's Laura Schulz joins experts from Hewlett Packard Enterprise and Intersect360 Research for "The next leap: Innovating with quantum, agentic AI, and HPC." The panel will explore how these technologies are converging to accelerate scientific breakthroughs and tackle complex challenges across industries.

https://www.hpe.com/us/en/discover/lasvegas.html

06/16/2026

How can automated testing boost scientific software development?

Join us on June 24 to learn how GitLab CI and ReFrame can help teams build scalable, portable testing capabilities and support accelerated development on ALCF's Aurora and Polaris supercomputers.

https://www.alcf.anl.gov/events/boosting-productivity-and-enabling-accelerated-development-gitlab-ci-and-reframe

06/15/2026

Using the ALCF’s Polaris supercomputer, an Argonne National Laboratory-led team carried out large-scale simulations that revealed the internal structure of the pion in unprecedented detail. Their work is helping scientists better understand how quarks and gluons combine to form visible matter while providing valuable guidance for future nuclear physics experiments.

https://www.alcf.anl.gov/news/researchers-use-argonne-supercomputer-illuminate-subatomic-particle-helps-hold-matter-together

06/12/2026

⏳ Time is running out to submit a 2027 INCITE proposal!

Researchers have until Monday, June 15, to apply for access to U.S. Department of Energy's exascale supercomputers, Aurora at the ALCF and Frontier at the Oak Ridge Leadership Computing Facility.

Open to teams from academia, industry, and government labs, the INCITE program supports ambitious simulation, AI, and data-intensive research campaigns tackling grand challenges in science and engineering.

Learn more and apply: https://doeleadershipcomputing.org/call-for-proposals/

06/11/2026

With the first project awarded through the ALCF's new AI Program for Exploration (APEX), researchers from the University of Notre Dame will develop an AI framework to accelerate the analysis of large-scale computational fluid dynamics simulations. The team will use Argonne National Laboratory's Aurora exascale system to run high-fidelity simulations and populate a knowledge graph, while leveraging the ALCF AI Inference Service to deploy LLM-based agents that extract insights from fluid dynamics data across a range of complex turbulent and reacting flow regimes.

"When considering large-scale computations, time-to-science is increasingly limited not by how quickly we can generate data, but by the manual effort required to interpret it," said Shivam Barwey of the University of Notre Dame. "Through APEX, we are working towards developing an AI framework that can reliably automate this analysis step for extreme-scale data, so we can accelerate the process of turning massive CFD simulations into a set of verifiable scientific insights — directly in natural language."

https://www.alcf.anl.gov/news/alcf-launches-apex-program-project-using-ai-advance-fluid-dynamics-research

06/09/2026

From working through theoretical physics problems to analyzing large simulation and experimental datasets, AI models are becoming an increasingly powerful tool for scientific discovery.

The ALCF Inference Service provides researchers with access to a range of large language models and foundation models running on Argonne National Laboratory's high-performance computing systems. A key enabler for DOE's Genesis Mission, the service helps researchers spend less time managing AI models and more time testing hypotheses, refining experiments, and extracting insights from complex data.

Learn more: https://www.alcf.anl.gov/news/alcf-launches-first-large-scale-ai-inference-service-open-science

ALCF user guide: https://docs.alcf.anl.gov/services/inference-endpoints/

ALCF webinar video: https://www.youtube.com/watch?v=22caQwCemug

06/08/2026

Congratulations to the AI for Materials team, which included several Argonne National Laboratory researchers, on receiving the Royal Society of Chemistry’s Materials Chemistry Horizon Prize! Led by longtime ALCF user Jacqueline Cole of the University of Cambridge, the collaboration leveraged ALCF supercomputers, AI, and experimental methods to accelerate the discovery of promising solar materials and develop open-source tools for the scientific community.

Learn more: https://www.anl.gov/article/argonnes-awardwinning-ai-tools-reshape-materials-design