Home

Building AI foundation models to accelerate the discovery of new battery materials

Science
A University of Michigan-led team is using ALCF supercomputers to develop massive foundation models that aim to revolutionize battery materials research.
Diagram of foundation model for battery materials

ALCF Hands-on HPC Workshop: Register by Sept. 5

Training
Join us for the 2025 ALCF Hands-on HPC Workshop, a two-part training event designed to help researchers boost the performance of HPC and AI applications and workflows on ALCF systems. The deadline to register is September 5, 2025.
alcf hands on workshop graphic featuring title and dates
Aurora Updates

Aurora Science and User News

Building AI Foundation Models to Accelerate the Discovery of New Battery Materials

HPCwire

Foundation Models to Accelerate the Discovery of New Battery Materials

CleanTechnica

US supercomputers train on billions of molecules to boost battery material discovery

Interesting Engineering

From Trust to Scale in LLMs: Insights from Key Leaders at TPC25

HPCwire

Learn More About Aurora

Improving next-generation aircraft with supercomputing and AI

Science
Using ALCF's Aurora exascale system, a University of Colorado Boulder-led team is combining large-scale simulations and machine learning to advance turbulence modeling and guide the design of more efficient aircraft.
Simulation showing airflow patterns around an airplane’s vertical tail.

DAOS and Lustre excel in MLPerf Storage Benchmark

Aurora
Aurora's DAOS and Lustre file systems achieved strong results in the MLPerf Storage Benchmark, demonstrating the capabilities needed for large-scale AI training in the exascale era.
MLPerf Storage

Researchers zero in on a new material for quantum information storage

Science
With help from ALCF’s Polaris supercomputer, the team performed high-precision calculations to shed light on a spin defect in magnesium oxide with potential for use as a qubit in future quantum systems.
Magnesium atoms (orange) and oxygen atoms (red) surround the nitrogen-vacancy center in magnesium oxide, shown by a transparent representation of a nitrogen atom under the missing magnesium atom. The yellow and blue spots show how electrons localize around the vacancy.

Aurora's dedication highlights advances in supercomputing and AI-driven research

Aurora
DOE, Argonne, and partners gathered to celebrate the lab's new Aurora exascale supercomputer and its role in accelerating scientific breakthroughs with powerful AI, simulation, and data analysis capabilities.
Aurora ribbon cutting

User Updates

Methods, Tools, and Best Practices for Coupling Simulation and AI on ALCF Systems

Webinar
Join us on August 27 for a webinar that will cover methods and tools to help scientists overcome challenges in integrating traditional HPC simulations with AI workloads on ALCF systems, using real-world examples to demonstrate best practices and effective workflows.
alcf dev session graphic
Events

Upcoming Training and Events

Democratizing HPC I/O Insights Using LLM Agents

08/26/2025

Methods, Tools, and Best Practices for Coupling Simulation and AI on ALCF Systems

08/27/2025

ParaView - Scientific Data Analysis and Visualization Training

10/02/2025

2025 ALCF Hands-on HPC Workshop

10/07 - 09/2025

See All Events
Updates

Recent Facility Updates

Minimum Node Size Extended for Aurora Tiny Queues

Aurora COE Office Hours: Tuesdays at 12 p.m. (CDT)

Maintenance Notice: Argonne Campus Network: August 22

Apply for a DD Allocation on Aurora

Extended Maintenance on Eagle, Polaris, Sophia, and Crux: Starting August 25

MobilePass+ (Cryptocard) Server Maintenance: August 26

See All Facility Updates

Machine Status

Mouse over diagrams for details

Aurora

  • Jobs —
  • Queued —
  • Reserved —
  • Nodes —
  • Usage —
 

Polaris

  • Jobs —
  • Queued —
  • Reserved —
  • Nodes —
  • Usage —
 

Sophia

  • Jobs —
  • Queued —
  • Reserved —
  • Nodes —
  • Usage —
 

Crux

  • Jobs —
  • Queued —
  • Reserved —
  • Nodes —
  • Usage —
 

See All Machine Status