Core42, a G42 company specializing in sovereign cloud and AI infrastructure, announced that its Maximus-384 supercomputer has been ranked No. 20 globally on the TOP500 List, the industry’s most recognized benchmark for publicly known supercomputing systems. The ranking highlights the performance of the AMD Instinct™ MI300X GPU-based “Maximus” cluster, operated at the company’s facility in Buffalo, New York and released ahead of Supercomputing 2025, the leading global conference for HPC.
The TOP500 list tracks performance trends across the global supercomputing systems using the High Performance Linpack benchmark (HPL), benchmark. Securing a top-20 position reflects exceptional engineering across compute, memory, storage, and networking, and validates the cluster’s ability to support large scale AI and HPC workloads under sustained production conditions.
Through G42’s strategic collaboration with AMD, Core42’s Buffalo deployment features more than 9,000+ AMD Instinct MI300X GPUs configured for large-scale AI training and inference. Delivered globally via the Core42 AI Cloud, the cluster provides organizations with access to advanced GPU capacity supported by unified orchestration, strong security controls, and comprehensive data management.
The Core42 AI Cloud infrastructure has been purpose-built as a high-performance computing (HPC) system optimized for AI workloads. Its compute island operates as a self-contained environment with dedicated high-performance storage and a backbone network delivering exceptional speed and reliability. This segmentation supports the ability to deploy multiple types of silicon which is central to the heterogeneous AI Cloud which Core42 has built which also includes Nvidia and Cerebras accelerators. This architecture is powered by Broadcom’s high-bandwidth networking chips, which handle data transfer between GPUs, CPUs, and storage, and Arista Networks’ Ethernet switches and EOS software, which manage traffic flow and optimize data routing.
The Core42 Maximus-384 cluster provides organizations with access to advanced GPU capacity supported by unified orchestration, enterprise-grade security, and comprehensive data-management controls. Its placement on the TOP500 List represents an important milestone for the Core42 AI Cloud and reinforces the company’s ability to operate enterprise-grade, high-performance environments at global scale.
Commenting on the achievement, Edmondo Orlotti, Chief Growth Officer at Core42, said:
“This ranking marks a new milestone in our mission to support the world’s most demanding workloads. Our expanding grid of AI infrastructure spans sovereign clusters and international supercomputing deployments, creating a global backbone built for ultra-low latency and regulatory compliance. Landing in the top 20 globally proves the quality of infrastructure we are building across regions and showcases the strength of our AI Cloud architecture and enterprise-grade capabilities. As AI adoption accelerates, we remain focused on expanding secure, high-performance environments that help organizations and individuals innovate with confidence. From startups to multinationals, we provide the tools to innovate responsibly and at scale.”
TOP500 says that the system has Linpack performance of 114.50 petaflops and a theoretical peak (Rpeak) of 251.15 petaflops.
The Buffalo cluster is operated within TeraWulf’s Lake Mariner facility as part of G42’s strategic partnership with the company. The site provides the power resiliency, cooling efficiency, and operational scale required for large GPU deployments, bringing together Core42’s high-performance AI infrastructure with TeraWulf’s zero-carbon, energy-efficient compute environment. As one of the international regions supporting the Core42 AI Cloud, the Buffalo facility enables scalable AMD Instinct MI300X GPU capacity for advanced training and inference workloads while maintaining the enterprise-grade performance, environmental standards, and reliability applied across every Core42 region.
“Core42’s Top500 achievement showcases the power of combining AMD Instinct MI300X GPUs with world-class AI infrastructure. Together, we’re advancing scalable, high-performance environments that enable customers to accelerate innovation and unlock the next wave of AI and HPC capabilities,” said Negin Oliver, corporate vice president, business development, Data Center GPU Business, AMD.
