# AMD powers 40% of TOP500 supercomputers as new Lux system debuts at rank 14

> **Key takeaway:** AMD has increased its presence in the global supercomputing rankings to 40%, bolstered by the debut of the Lux AI-focused supercomputer.

**Published:** 2026-11-16T21:00:00+00:00 
**Source:** Tom&#x27;s Hardware 
**Category:** open-hardware 
**Canonical URL:** https://fosswire.org/news/amd-powers-40-of-top500-supercomputers-as-new-lux-system-deb.html 

## Summary
The latest TOP500 list reveals a significant expansion of AMD's footprint in high-performance computing, with the company now powering 202 systems. This represents a 14% increase in AMD-based machines compared to the 2025 rankings, bringing their total market share of the list to over 40%.

A major contributor to this growth is the debut of Lux, a new supercomputer ranked 14th. Developed by AMD and the U.S. Department of Energy (DOE) in collaboration with HPE, Lux is designed specifically for AI research and scientific discovery under the Genesis Mission initiative.

## Technical specifications of the Lux supercomputer
- **GPU Architecture:** The Lux system utilizes over 4,000 AMD Instinct MI355X GPUs distributed across more than 500 nodes.
- **Memory Capacity:** Each GPU is equipped with 288 GB of HBM3E memory, providing the entire system with a total of 1.1 PB of high-bandwidth memory.
- **Compute Performance:** The machine delivers 317 PetaFLOPS for FP64 HPC workloads, 20 ExaFLOPS of FP8 for AI training, and 40 ExaFLOPS of FP4 for AI inference.
- **System Integration:** Nodes are built within HPE ProLiant Compute XD685 chassis, utilizing AMD EPYC host CPUs and AMD Pensando networking hardware.

## Why it matters
The rise of AMD in the TOP500 highlights the growing importance of open-standard hardware architectures in large-scale AI and scientific computing. For the Linux ecosystem, the deployment of these massive clusters relies heavily on specialized distributions and high-performance networking stacks to manage exascale workloads.