Washington D.C. – NVIDIA and Oracle have announced a landmark collaboration with the U.S. Department of Energy (DOE) and Argonne National Laboratory to construct the nation’s largest artificial intelligence supercomputer, a strategic move aimed at dramatically accelerating scientific discovery and reinforcing U.S. leadership in AI. The initiative, unveiled during NVIDIA’s GTC conference, will see the development of two powerful systems, Solstice and Equinox, at Argonne National Laboratory in Illinois.
The Solstice system alone will feature a groundbreaking 100,000 NVIDIA Blackwell GPUs, making it the largest AI supercomputer within the DOE’s laboratory complex.
Powering Next-Generation Scientific Discovery
This public-private partnership is designed to equip researchers with unparalleled computational capabilities to tackle some of the nation’s most complex challenges. The Solstice and a companion system, Equinox, which will incorporate 10,000 NVIDIA Blackwell GPUs and is expected in the first half of 2026, will together deliver a combined 2,200 exaflops of AI computing performance.
These supercomputers are intended to revolutionize research and development across various fields, including healthcare, materials science, clean energy, fusion research, molecular modeling, and national security applications.
The Role of NVIDIA Blackwell GPUs and Oracle Cloud Infrastructure
At the core of these new systems are NVIDIA’s cutting-edge Blackwell GPUs, designed for massively parallel workloads and deep learning at an unprecedented scale. NVIDIA founder and CEO Jensen Huang emphasized the significance, stating, “AI is the most powerful technology of our time, and science is its greatest frontier.” He added that together with Oracle, they are building “America’s engine for discovery, giving researchers access to the most advanced AI infrastructure to drive progress across fields ranging from healthcare research to materials science.”
The collaboration also leverages Oracle Cloud Infrastructure (OCI) to provide immediate access to AI computing resources for DOE researchers, utilizing a combination of NVIDIA Hopper and Blackwell architectures. This immediate access is part of a phased approach to swiftly empower scientific endeavors.
Strategic Importance and Public-Private Partnership Model
U.S. Secretary of Energy Chris Wright highlighted the strategic imperative of such alliances. “Winning the AI race requires new and creative partnerships that will bring together the brightest minds and industries American technology and science has to offer,” Wright said. This initiative reflects a new public-private partnership model for the DOE, encouraging industry investments and shared computing power between government and industry.
The Solstice and Equinox systems will enable scientists to develop and train new frontier models and AI reasoning models for open science using the NVIDIA Megatron-Core library and scale them with NVIDIA’s TensorRT™ inference software stack. These models will form the backbone of “agentic AI workflows,” aiming to boost R&D productivity and accelerate discovery within a decade.
Integration with Existing Facilities and Broader Impact
The new AI supercomputers at Argonne National Laboratory will seamlessly connect with the DOE’s extensive network of scientific instruments and data assets, including experimental facilities like the Advanced Photon Source. This integration is expected to create real-time feedback loops between data collection and AI-driven insights, drastically shortening the path from hypothesis to breakthrough in fields such as materials design and pharmaceuticals.
Beyond Solstice and Equinox, NVIDIA is also involved in building several other AI supercomputers for the Department of Energy across Argonne and Los Alamos national laboratories. These include systems like Tara, Minerva, and Janus at Argonne, and the Vera Rubin platform for the Mission and Vision systems at Los Alamos, further solidifying the nation’s AI infrastructure.
This monumental partnership between NVIDIA, Oracle, and the U.S. Department of Energy signifies a critical step in the global race for AI leadership, ensuring that America maintains a robust foundation for future innovation and scientific advancement.
