San Jose, CA – March 17, 2026 – MSI, a global leader in high-performance server solutions, today unveils its latest AI infrastructure portfolio built on NVIDIA’s modular architectures, including the NVIDIA MGX platform and NVIDIA DGX Station technology. Designed to accelerate AI training, large-scale inference, HPC, edge, and next-generation data center workloads, MSI’s expanded lineup delivers exceptional scalability, performance density, and deployment flexibility.

Leveraging the modular design of NVIDIA MGX architecture, MSI has developed a comprehensive portfolio of 4U and 6U liquid-cooled servers supporting NVIDIA RTX PRO 6000 Blackwell Server Edition and NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs. The NVIDIA MGX architecture enables flexible CPU selection, high-capacity memory configurations, and seamless integration of high-speed networking — empowering enterprises to deploy infrastructure tailored to diverse workload requirements, from data center deployments to edge applications.

MSI’s platforms based on NVIDIA MGX support both Intel and AMD CPU options, up to 32 DDR5 DIMM slots, and as many as eight 400G Ethernet ports powered by NVIDIA ConnectX-8 SuperNICs. With advanced thermal engineering, including both liquid cooling and optimized air-cooling designs, these systems are purpose-built to sustain peak performance under the most demanding AI workloads.

The CG480-S5063 represents MSI’s flagship 4U server based on NVIDIA MGX, featuring dual Intel® Xeon® 6 processors, eight dual-width PCIe GPU slots, and 32 DDR5 DIMM slots for maximum memory scalability. Supporting up to twenty PCIe Gen5 E1.S NVMe bays, the system delivers ultra-fast data throughput and expansive storage capacity.

Expanding the NVIDIA MGX family, the CG481-S6053 integrates dual AMD EPYC™ 9005 Series processors to maximize core density and I/O bandwidth. Featuring eight PCIe 5.0 GPU slots, 24 DDR5 DIMM slots and eight high-bandwidth 400G Ethernet ports via NVIDIA ConnectX-8 SuperNICs, this platform is designed for compute-intensive AI clusters, HPC simulations, and multi-tenant enterprise AI deployments.

The CG681-S6093 is a 6U liquid-cooled AI platform built on a dual-socket architecture, supporting eight dual-width PCIe GPU slots and eight 400G Ethernet ports powered by NVIDIA ConnectX-8 SuperNICs. It delivers exceptional performance and efficiency for high-density AI data center deployments.

To reinforce its commitment to next-generation AI computing, MSI is showcasing NVIDIA’s Vera CPU option, highlighting its ongoing innovation in data-driven, AI-powered infrastructure solutions.

The NVIDIA MGX architecture is particularly well-suited for computer vision and real-time video analytics applications. With high GPU throughput and ultra-fast networking, MSI’s AI platforms enable real-time multi-camera processing for smart cities, industrial inspection, and advanced surveillance systems.

At GTC 2026, MSI is demonstrating AI-powered smart video search and automated summarization capabilities, showcasing how enterprises can extract actionable intelligence from massive volumes of live and archived video data.

For AI developers, researchers, and data scientists requiring data center-class performance in a workstation form factor, MSI announces the availability of the XpertStation WS300 beginning March 16.

Built upon the NVIDIA DGX Station architecture, the WS300 is powered by NVIDIA Grace Blackwell Ultra Desktop Superchip and features 748GB of large coherent memory. Equipped with dual 400GbE networking ports via NVIDIA ConnectX-8 SuperNICs and a robust 1600W ATX power supply, the system delivers unprecedented AI compute capability directly at the deskside with a plug-and-play supercomputing feature.

The XpertStation WS300 enables advanced AI model development, fine-tuning, inference, data science workflows, and complex simulations — bringing data center-level acceleration to enterprise developers and research teams.

With its expanded NVIDIA MGX server portfolio and NVIDIA DGX-powered workstation platform, MSI continues to deliver scalable, high-performance AI infrastructure, empowering organizations to accelerate innovation from the data center to the edge and the desktop.