Supermicro Launches AI Systems Powered by NVIDIA Blackwell Ultra
Supermicro, Inc. (NASDAQ: SMCI), a Total IT Solution Provider for AI, Cloud, Storage, and 5G/Edge, is announcing new systems and rack solutions powered by NVIDIA’s Blackwell Ultra platform, featuring the NVIDIA HGX B300 NVL16 and NVIDIA GB300 NVL72 platforms. Supermicro and NVIDIA’s new AI solutions strengthen leadership in AI by delivering breakthrough performance for the most compute-intensive AI workloads, including AI reasoning, agentic AI, and video inference applications.
Supermicro’s Blackwell Ultra-powered solutions are engineered to address the demanding requirements of modern AI workloads. The NVIDIA Blackwell Ultra platform delivers unprecedented performance for AI training and inference. This includes significant improvements in AI FLOPS, memory capacity, and network bandwidth. The company’s approach leverages the NVIDIA HGX B300 NVL16 and GB300 NVL72 platforms to provide customers with a streamlined path to deploying cutting-edge AI solutions.
The NVIDIA Blackwell Ultra platform is built to conquer the most demanding cluster-scale AI applications by overcoming performance bottlenecks caused by limited GPU memory capacity and network bandwidth. NVIDIA Blackwell Ultra delivers an unprecedented 288GB HBM3e of memory per GPU, delivering drastic improvements in AI FLOPS for AI training and inference for the largest AI models. The networking platform integration with NVIDIA Quantum-X800 InfiniBand and Spectrum-Xâ„¢ Ethernet doubles the compute fabric bandwidth, up to 800 Gb/s.
Supermicro integrates NVIDIA Blackwell Ultra with two types of solutions: Supermicro NVIDIA HGX B300 NVL16 systems, designed for every data center, and the NVIDIA GB300 NVL72, equipped with NVIDIA’s next-generation Grace Blackwell architecture. Supermicro’s HGX B300 NVL16 system is a standard building block for AI training clusters, featuring an 8-GPU NVIDIA NVLink™ domain and a 1:1 GPU-to-NIC ratio for high-performance clusters. The NVIDIA GB300 NVL72 integrates 72 NVIDIA Blackwell Ultra GPUs and 36 NVIDIA Grace™ CPUs in a single rack with exascale computing capacity.
Supermicro’s approach focuses on reducing power consumption and mitigating environmental impact, a cornerstone of their Server Building Block Solutions® portfolio. Specifically, the HGX B300 NVL16 system utilizes a liquid-cooled solution with a 250kW in-rack CDU and cooling tower. This innovative approach allows for operating temperatures as low as 35℃ in a 16-node rack configuration, significantly reducing power consumption compared to air-cooled systems. Furthermore, the system supports up to 40% power savings.
Supermicro’s expertise in liquid cooling, data center deployment, and building block approach positions them to deliver NVIDIA Blackwell Ultra with industry-leading time-to-deployment. The company offers a complete liquid cooling portfolio, including newly developed direct-to-chip cold plates. Supermicro’s on-site rack deployment helps enterprises build data centers from the ground up, including planning, design, power-up, validation, testing, installation, and configuration of racks, servers, switches, and other networking equipment to meet the organization’s specific needs. This comprehensive support ensures seamless integration and optimal performance.
Supermicro will be exhibiting its new systems at GTC 2025 in San Jose, CA, from March 17-21, 2025. Attendees are encouraged to visit booth #1115 to see the X14/H14 B200, B300, and GB300 systems on display along with their rack-scaled liquid-cooled solutions. Supermicro’s commitment to innovation and collaboration with NVIDIA underscores its position as a leader in the rapidly evolving AI landscape.
