GIGABYTE Technology has expanded its end-to-end AI infrastructure portfolio, designed specifically for the telecommunications industry. GIGABYTE’s end-to-end solutions, unveiled at MWC 2026 like that of other competitors such as MSI, enable operators to convert massive volumes of network data into intelligence, automation and new revenue streams, as telecom networks evolve from data carriers to AI-powered digital platforms.
At the center of the transformation from telecommunications to AI is AI Factorywhere network and subscriber data is converted into operational intelligence and commercialized AI services. GIGABYTE addresses this need with the GB300 NVL72, a liquid-cooled rack-scale platform that integrates 72 NVIDIA Blackwell Ultra GPUs and 36 NVIDIA Grace CPUs into a single system.
Connected via NVIDIA Quantum-X800 InfiniBand or NVIDIA Spectrum-XTM Ethernet and NVIDIA ConnectX-8 SuperNICTM, GB300 NVL72 is optimized for large-scale AI training and inference workloads, enabling operators to automate operations, optimize network planning, and deploy AI-powered services at telecom scale.
Driving the data chain to AI with AI and HPC supercomputing
To eliminate bottlenecks in the process of data to AIGIGABYTE expands its AI and HPC portfolio to the telecommunications sector with platforms based on NVIDIA and AMD accelerator architectures. The G894-SD3-AAX7, powered by NVIDIA HGX B300, supports workloads such as real-time traffic analysis, reasoning models, and large-scale AI training. For converged AI and HPC environments, the
Digital twins for AI-powered intelligent network operations
Digital twins are becoming a crucial tool for AI-based network operations. GIGABYTE enables real-time high-fidelity simulation with XL44-SX2-AAS1, based on the NVIDIA MGX architecture and configured with eight NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs. With 800 GB/s bandwidth via NVIDIA ConnectX-8 SuperNICTM and PCIe Gen6 connectivity, the platform enables higher performance and scale-out capabilities, making it ideal for 24/7 high availability tasks.
Extending AI from the core to the edge
The AI transformation extends to the edge of the network, where low latency, data proximity and privacy are essential. GIGABYTE AI Workstations enable local AI development, inference and deployment in enterprise and private network environments. The W775-V10-L01, powered by the NVIDIA GB300 superchip, supports up to 775GB of coherent memory for large-scale AI workloads on the desktop. Additional platforms, including AMD EPYC and Intel Xeon-based workstations, provide flexibility for various edge AI and private networking scenarios. Delivering up to a petaFLOP of AI computing, the compact AI TOP ATOM further enables rapid prototyping and deployment of edge AI in a palm-sized form factor.
