Projects

As part of DOE's broader Energy Exascale Earth System Model (E3SM) project, researchers are using exascale supercomputers to evaluate the SCREAM model through real-world simulations at an unprecedented 200-meter resolution.

This project will use DOE exascale computing and decades of VA health record data to build long-context clinical LLMs that can better model patient histories, support suicide risk and frailty progression analysis, and provide explainable insights for VA clinicians.

This project will use exascale computing, AI, and multiscale simulation to advance human digital twins for personalized medicine, focusing on molecular-scale drug discovery and organ-scale hemodynamics.

The FRAME-IDP project uses generative AI, large-scale molecular simulations, and experimental validation to design biologics that can target intrinsically disordered proteins involved in cancer signaling, turning previously “undruggable” proteins into a new class of therapeutic opportunities.

This INCITE project uses exascale, full-core multiphysics simulations of advanced fission and fusion reactors to support the design and deployment of next-generation nuclear energy systems.

This INCITE project is using quantum-accurate simulations and machine-learned models to reveal atomic-scale interfacial reactions in batteries and catalysts to improve energy technology performance.

This INCITE project uses exascale molecular dynamics simulations and coordinated experiments to understand brittle and ductile failure in tungsten polycrystals and alloys, with the goal of improving additive manufacturing and fusion energy materials performance.

This INCITE project aims to build and evaluate large, scientifically trained and aligned foundation models and tools that integrate AI into research workflows to accelerate scientific discovery.

This INCITE project uses large-scale simulations and optimization to improve laser-wakefield acceleration, staged beam coupling, and strong-field photon production, supporting experiments and guiding the design of next-generation laser-based accelerators and radiation sources.

This INCITE project uses advanced simulations with upgraded laser models to study how complex, structured laser pulses interact with plasmas, aiming to enable new accelerator, light source, and fusion-related technologies.

The INCITE project leverages exascale computing, machine-learned surrogates, and advanced core–edge plasma simulations to accurately predict temperature, density, and heat-flux profiles in fusion devices, providing validated workflows to guide future tokamak design.

This project develops and scales BioM3, a multimodal deep generative foundation model that uses natural language prompts to design functional proteins, advancing fundamental understanding of protein mechanisms and enabling new applications in medicine, biotechnology, and engineering.

This project develops the first genome-scale foundation model capable of whole-chromosome analysis across all domains of life, leveraging DOE’s exascale computing and genomic datasets to advance AI-driven discovery in bioenergy, environmental science, and national security.

This project develops a digital twin of a microfluidic device using Advanced Physics Refinement to enable high-throughput mechano-phenotyping and systematically quantify cellular behavior across realistic red blood cell configurations in whole blood.

AbacusAurora produces a 35-trillion-particle cosmological N-body simulation—supported by a companion parameter-sweep suite—to model large-scale structure with unprecedented volume and resolution, providing a foundational resource for next-generation surveys and studies of dark energy, dark matter, and galaxy formation.

This project uses exascale radiation-hydrodynamic simulations of tidal disruption events to model how supermassive black holes tear apart stars and shape luminous accretion flows, enabling interpretation of new data from the Rubin Observatory.

This project uses exascale, turbulence-resolving simulations and machine learning–enhanced design models to optimize integrated open fan propulsion systems, enabling step-change improvements in aircraft fuel efficiency and emissions reduction.

This project uses fully resolved direct numerical simulations to study and quantify how turbulence, bubbles, and scalar diffusion interact in gravity-driven bubbly flows. The goal is to improve predictive understanding of mixing, energy transport, and fluxes in turbulent bubbly suspensions to advance high-performance engineering applications.

This project uses ultra-high-resolution, general-relativistic simulations of neutron star mergers to improve gravitational-wave modeling and interpret multimessenger observations from LIGO and its international partners, advancing understanding of extreme gravity, heavy-element formation, and the engines of gamma-ray bursts and kilonovae.

The INCITE proposal will generate a massive high-resolution simulation dataset of shock wave–boundary layer interactions using advanced supercomputers to better understand turbulence in hypersonic flows. The results will support improved turbulence models, hypersonic vehicle design, and future machine-learning–based CFD methods, with the data shared publicly for research use.