BARCELONA, Spain – March 2, 2026 – MSI, a leading global provider of high-performance server solutions, is showcasing its latest AI-vRAN solutions at MWC 2026 (Booth #5A61) to address the evolving requirements of next-generation telecom networks. MSI’s AI-vRAN stack, integrated with GPU server solutions, is designed to bring AI capabilities into the core of network operations, supporting growing telecommunications workloads and accelerating the transition toward AI-powered mobile networks.

“AI is rapidly reshaping the telecommunications landscape, and MSI is focused on helping operators build scalable network architectures that support diverse applications, including voice, data, video streaming, and AI workloads, through a unified infrastructure,” said Danny Hsu, General Manager of Enterprise Platform Solutions at MSI. “This AI-powered vRAN architecture delivers greater flexibility for 5G and future 6G deployments while accelerating time-to-market for new services.”

To support AI-driven telecom networks, MSI delivers a unified AI-vRAN platform designed to simplify deployment across O-RAN, private 5G, and virtual RAN environments. This unified architecture enables operators to adopt AI-powered network functions while maintaining consistency across distributed and centralized network infrastructures.

At MWC 2026, MSI highlights its CG480-S6053 and CG290-S3063 platforms, purpose-built to address diverse AI-vRAN deployment requirements. The CG480-S6053 is designed to support dense GPU configurations for compute-intensive AI inference and acceleration workloads. Meanwhile, the CG290-S3063 is optimized for space and power-efficient deployments in a 2U chassis, making it well suited for edge and distributed network environments. Both platforms offer flexible configurations of GPUs, NICs, DPUs, and storage to help operators optimize performance, scalability, and cost efficiency across a consistent infrastructure.

Alongside these systems, the 2U CX271-S4056 platform (HE SKU) is designed to support balanced data processing for AI workloads, addressing scenarios where compute efficiency and system balance are critical.

A key capability of MSI’s AI-vRAN solution is dynamic GPU allocation, which enables resources to be flexibly assigned between 5G communication workloads and AI workloads based on real-time demand. By supporting both Base Station and Edge AI functions within the same infrastructure, the platform enables simultaneous processing of RAN and AI workloads while improving overall compute utilization. Integrated software support further ensures AI services can be efficiently deployed and operated at the network edge.

MSI offers a scalable range of GPU-accelerated server configurations, from 2 to 8 GPUs, to address diverse deployment requirements across wireless access, metro edge, core network, and centralized cloud environments. This flexibility allows telecom operators to align system configurations with specific workload demands while optimizing performance and operational efficiency.

Built on the NVIDIA MGX architecture, MSI’s AI platforms have demonstrated reliability and scalability through widespread adoption in AI data centers worldwide. MSI is now extending this proven architectural approach into the telecom domain, applying data center–class computing principles to support the evolving performance, scalability, and efficiency requirements of next-generation mobile networks.

Supporting Resources:
Watch the MSI’s 4U & 2U NVIDIA MGX AI platform, built on NVIDIA accelerated computing to deliver the performance for tomorrow’s AI and telecom workloads.