Newswise — A new computing platform that combines artificial intelligence (AI) with high-performance computing aims to end the bottleneck holding back fusion energy research by accelerating the simulations needed to advance the field.
The project – known as Simulation, Technology and Experimentation Leveraging Research Accelerated by AI-Enabled Learning (STELLAR-AI) – will be led by the(DOE) Princeton Plasma Physics Laboratory (PPPL). However, STELLAR-AI will extend far beyond the laboratory walls, bringing together national laboratories, universities, technology companies and industry partners to build the computing foundation of needs of the community.
Running a single high-fidelity computer simulation or training an artificial intelligence (AI) system that can design an ideal fusion system using existing infrastructure can take months. STELLAR-AI is designed to reduce this delay using artificial intelligence. The platform connects computing resources directly to experimental devices, including PPPLs. Upgrading the national experience on the spherical torus (NSTX-U), which is scheduled to go live this year, allowing researchers to analyze data as experiments are carried out.
Building the IT base for the merger
Jonathan Menarddeputy director of research at PPPL, views STELLAR-AI as the cornerstone of the U.S. fusion ecosystem: a dedicated, AI-driven research environment built specifically for the fusion energy mission. STELLAR-AI will combine speed and precision, accelerating the path to commercially viable fusion energy.
“Fusion is a complex system of systems. We need AI and high-performance computing to really optimize the design for economical construction and operation,” Ménard said. “We want to combine simulation technology and experiments – particularly NSTX-U – with AI and partnerships to accelerate fusion. »
STELLAR-AI will achieve this goal by integrating CPUs, GPUs, and QPUs into an ideal hardware configuration to address the challenges private fusion companies face as they strive to bring a solution to market. Processors, or central processing units, are standard computer chips that handle everyday computing tasks. GPUs, or graphics processing units, are specialized chips that excel at the parallel calculations needed for artificial intelligence. QPUs, or quantum processing units, use the principles of quantum physics to solve certain complex problems that would take much longer for traditional computers to solve.
An essential part of the Genesis mission
STELLAR-AI is part of the a national effort launched by in November 2025 to use AI to accelerate scientific discovery in DOE laboratories.
“The Genesis Platform is an integrated and ambitious system that will bring together DOE’s various unique assets: experimental and user facilities, supercomputers, data archives and, most importantly, AI models,” said Shantenu Jhahead of the computational sciences department at PPPL. While Genesis provides this vast infrastructure, STELLAR-AI brings fusion-specific computer codes, data and scientific models into the national system. The project also aligns with DOE recommendations which calls for building an AI-Fusion digital convergence platform to accelerate the commercialization of a fusion power plant, achieve U.S. energy dominance, and provide the abundant power needed to drive the next generation of AI and computing.
The researchers plan to use STELLAR-AI for projects spanning simulation, design, and support of real-time experiments. One effort will aim to create a digital twin of NSTX-U: a computer model that mirrors the physical machine so closely that scientists can test their ideas virtually before launching real experiments. Another project, called StellFoundry, uses AI to accelerate the design of stellarators, a type of fusion device with a twisted, pretzel-like shape that some scientists say could offer advantages over other designs. Designing Stellarator requires sifting through massive amounts of data to find the best configurations, a process that traditionally takes months or years and will greatly benefit from the STELLAR-AI platform.
A network of public and private partners
STELLAR-AI’s strength lies in PPPL’s partnerships with DOE national laboratories, AI and HPC companies, academic institutions, and fusion and engineering companies. The team includes world-leading capabilities from national laboratories including PPPL and as well as top universities such as And . which manages the laboratory on behalf of the U.S. DOE Office of Science, is also a key partner. Princeton will support operations, research software engineering, and user training for the STELLAR-AI infrastructure. Crucial technical support comes from tech giants like which provides expertise to improve the performance of several critical merge codes, and which will bring together the main cloud functionalities of Azure. We also work directly with the fusion industry, including , , And . This unique combination of partners will provide proven AI models and key tools for the U.S. fusion industry.
STELLAR-AI is just one of several initiatives which position PPPL as a hub for public-private collaboration in the field of fusion energy. The laboratory’s seven decades of plasma research, combined with experimental facilities such as NSTX-U and computing expertise, have made it a destination for companies and research institutions seeking to accelerate fusion development.
