_(1).jpg)
SuperX AI Technology Limited, a Singapore-based AI infrastructure solutions provider, has launched its flagship XN9160-B300 AI Server, powered by NVIDIA’s Blackwell B300 GPUs. Engineered for high-performance computing, this server addresses the growing demand for scalable AI training, machine learning, and HPC applications, offering unmatched performance for enterprise-scale environments. For stock price details, see the finance card above.
The XN9160-B300 is purpose-built for intensive AI workloads, including foundation model training, reinforcement learning, and multimodal AI models. Its 8 NVIDIA Blackwell B300 GPUs, supported by the HGX B300 module, deliver 50% more NVFP4 compute and HBM3E memory capacity per chip compared to standard Blackwell, enabling faster training and inference for large-scale models. The server also excels in HPC applications such as drug discovery, seismic analysis, and climate modeling, making it ideal for research and enterprise use.
Designed for hyperscale AI factories, the XN9160-B300 integrates eight 800Gb/s OSFP ports for InfiniBand or dual 400Gb/s Ethernet, ensuring low-latency communication for AI SuperPOD clusters. The fifth-generation NVLink interconnects allow the GPUs to function as a single accelerator, supported by 2,304GB of unified HBM3E memory (288GB per GPU). This massive memory pool eliminates offloading delays, supporting high-concurrency generative AI and large language models with extensive Key/Value caches.
The server’s dual Intel Xeon 6 processors and 32 DDR5 DIMMs (up to 8000MT/s) provide the bandwidth to keep pace with GPU demands. Twelve 3000W 80 PLUS Titanium power supplies ensure energy efficiency and reliability under continuous peak loads. Additionally, eight 2.5" Gen5 NVMe hot-swap bays and multiple PCIe Gen5 x16 slots offer comprehensive storage and expansion options, making the XN9160-B300 a compact yet powerful solution for mission-critical data centers.
The XN9160-B300 targets diverse sectors: hyperscale AI for cloud providers, scientific simulations for research institutions, financial services for real-time risk modeling, bioinformatics for genomics, and global systems modeling for climate and disaster prediction. Its scalable, energy-efficient architecture positions it as a cornerstone for organizations pushing AI and HPC boundaries.
This launch marks a pivotal step in SuperX’s AI infrastructure roadmap, delivering cutting-edge compute capabilities to drive global innovation. By combining NVIDIA’s Blackwell Ultra technology with enterprise-ready design, the XN9160-B300 empowers organizations to accelerate AI development and achieve transformative outcomes across industries.
Super X AI Technology Limited is an AI infrastructure solutions provider, offering a comprehensive portfolio of proprietary hardware, advanced software, and end-to-end services for AI data centers. The Company's services include advanced solution design and planning, cost-effective infrastructure product integration, and end-to-end operations and maintenance. Its core products include high-performance AI servers, High-Voltage Direct Current (HVDC) solutions, high-density liquid cooling solutions, as well as AI cloud and AI agents. Headquartered in Singapore, the Company serves institutional clients globally, including enterprises, research institutions, and cloud and edge computing deployments.
TechIntelPro empowers B2B technology decision-makers with the latest industry trends, deep insights, leadership perspectives and actionable intelligence to drive both organizational and professional milestones—and thrive on the tech pulse.