Supermicro Begins Volume Shipments of NVIDIA Blackwell Ultra Systems and Rack Plug-and-Play Data Center-Scale Solutions

Rhea-AI Summary

Supermicro (NASDAQ: SMCI) has announced volume shipments of its NVIDIA Blackwell Ultra solutions, including NVIDIA HGX B300 systems and GB300 NVL72 racks worldwide. The company's Data Center Building Block Solutions (DCBBS) offers pre-validated, plug-and-play solutions at system, rack, and data center scale.

Key performance metrics include 1.1 exaFLOPS dense FP4 compute performance for GB300 NVL72 rack-scale systems and up to 7.5x performance gains over NVIDIA Hopper systems. The solutions feature 800 Gb/s bandwidth, 144 petaFLOPS of FP4 compute, and 270 GB of HBM3e memory per GPU.

Through DCBBS implementation, customers can achieve 40% power savings, 60% reduced data center footprint, 40% decreased water consumption, and 20% lower TCO.

Positive

• New systems deliver 7.5x performance gains over previous NVIDIA Hopper systems
• Achieves 1.1 exaFLOPS dense FP4 compute performance in rack-scale systems
• Solutions enable 40% power savings and 20% lower Total Cost of Ownership
• 50% greater inferencing performance and 50% more HBM3e capacity compared to NVIDIA Blackwell
• Pre-validated, plug-and-play solutions enable rapid deployment and faster time-to-market

Negative

• Higher power requirements of up to 1400W per GPU may increase operational costs
• Complex infrastructure requirements including specialized cooling and networking systems

News Market Reaction

+2.39%

36 alerts

+2.39% News Effect

+2.7% Peak in 16 hr 17 min

+$610M Valuation Impact

$26.12B Market Cap

0.9x Rel. Volume

On the day this news was published, SMCI gained 2.39%, reflecting a moderate positive market reaction. Argus tracked a peak move of +2.7% during that session. Our momentum scanner triggered 36 alerts that day, indicating elevated trading interest and price volatility. This price movement added approximately $610M to the company's valuation, bringing the market cap to $26.12B at that time.

Data tracked by StockTitan Argus on the day of publication.

• Now delivering high volume NVIDIA HGX B300 systems and NVIDIA GB300 NVL72 in volume to customers worldwide
• Enabling turn-key day-one operation by delivering pre-validated solutions at system, rack, and data center-scale through Supermicro Data Center Building Block Solutions^® (DCBBS)
• Complete portfolio of 10+ SKUs now available featuring NVIDIA Blackwell and Blackwell Ultra to create diverse AI Factory environments at any scale

SAN JOSE, Calif., Sept. 11, 2025 /PRNewswire/ -- Supermicro, Inc. (NASDAQ: SMCI), a Total IT Solution Provider for AI, Cloud, Storage, and 5G/Edge, today announced the broad availability of its NVIDIA Blackwell Ultra solutions. Supermicro is now delivering Plug-and-Play (PnP)-ready NVIDIA HGX B300 systems and GB300 NVL72 racks to customers worldwide. These solutions are purpose-built and pre-validated at system, rack, and data center scale before shipping, enabling rapid deployment of the industry's highest performance and compute density for transformative AI infrastructure spanning the largest-scale AI training, real-time AI reasoning, agentic AI applications, multimodal AI inference, and physical AI deployments.

"Supermicro has the best track record of fast and successful deployments of new NVIDIA technologies," said Charles Liang, president and CEO of Supermicro. "Through Supermicro Data Center Building Block Solutions with our expertise in on-site deployment, we enable turn-key delivery of the highest-performance AI platform — critical for customers seeking to invest in cutting-edge technology. Data center customers face many AI infrastructure challenges: complex network topology and cabling, power delivery, and thermal management. Supermicro delivers pre-validated, plug-and-play solutions at system, rack, and data center scale, enabling AI factories to deploy rapidly and helping our customers lead in AI."

For more information, please visit https://www.supermicro.com/en/accelerators/nvidia

Supermicro integrates Blackwell Ultra generational advancements with system-level and rack-level engineering to create a solution that maximizes efficiency and performance at scale. At the system level, Supermicro's NVIDIA Blackwell Ultra systems feature advanced air and liquid cooling design optimized for the increased GPU power utilization. Blackwell Ultra's GB300 and B300 can be configured to use up to 1400W per GPU while offering 50% greater inferencing performance while utilizing FP4 compute and 50% more HBM3e capacity compared to NVIDIA Blackwell. Having larger memory sizes and faster inferencing speeds is crucial for running more complex and powerful models efficiently.

Supermicro's complete Blackwell Ultra portfolio leverages breakthrough innovations including the company's direct liquid cooling (DLC) technology stack, advanced air cooling, and I/O design optimization. Validated by market-proven deployments, Supermicro delivers at volume the industry's broadest portfolio of NVIDIA Blackwell systems. These Supermicro NVIDIA Blackwell Ultra solutions are specifically engineered as the building blocks for AI factories of the future and doubling compute network bandwidth to achieve new benchmarks in AI factory productivity.

Breakthroughs in AI, including the development of foundation models now with trillions of parameters, are forged at the cluster and data center-level with enormous AI factories interconnected with high-bandwidth networking. Supermicro offers NVIDIA Blackwell Ultra reference architecture solutions for enterprises seeking plug-and-play deployment of NVIDIA Quantum-X800 InfiniBand or NVIDIA Spectrum-X™ Ethernet compute fabrics with bandwidth up to 800 Gb/s. From a single rack to cluster level configurations as fully scalable plug-and-play units, with a choice of air or liquid cooling, these reference architecture-based solutions allow customers to take full advantage of Blackwell Ultra integration of 800 Gb/s NVIDIA ConnectX-8 SuperNICs into the NVIDIA GB300 NVL72 and NVIDIA HGX B300. Supermicro's GB300 NVL72 rack-scale system achieves 1.1 exaFLOPS dense FP4 compute performance, while NVIDIA HGX B300 systems in 8U air-cooled and 4U liquid-cooled configurations provide up to 7.5x performance gains over NVIDIA Hopper™ accelerator based systems, with 144 petaFLOPS of FP4 compute and 270 GB of HBM3e memory per GPU.

Supermicro's total solution with NVIDIA Blackwell Ultra is fully integrated to combine the hardware with infrastructure software and application software, including NVIDIA AI Enterprise, NVIDIA Blueprints, and NVIDIA NIM to bring optimized AI performance to these powerful systems. In addition, Supermicro's DCBBS offers a range of services, including on-site deployment of cluster cabling and other critical data center components, including power and thermal equipment, to ensure fast time-to-market and time-to-online. Combined with its DLC-2 (direct liquid cooling) technology, DCBBS helps customers save up to 40% power, reducing 60% data center footprint, and decreasing 40% water consumption, all of which leads to 20% lower TCO. As AI factories continue to scale, Supermicro's AI Factory DCBBS package fully equips data centers to tackle these rising AI computational requirements.

About Super Micro Computer, Inc.

Supermicro (NASDAQ: SMCI) is a global leader in Application-Optimized Total IT Solutions. Founded and operating in San Jose, California, Supermicro is committed to delivering first to market innovation for Enterprise, Cloud, AI, and 5G Telco/Edge IT Infrastructure. We are a Total IT Solutions provider with server, AI, storage, IoT, switch systems, software, and support services. Supermicro's motherboard, power, and chassis design expertise further enables our development and production, enabling next generation innovation from cloud to edge for our global customers. Our products are designed and manufactured in-house (in the US, Taiwan, and the Netherlands), leveraging global operations for scale and efficiency and optimized to improve TCO and reduce environmental impact (Green Computing). The award-winning portfolio of Server Building Block Solutions^® allows customers to optimize for their exact workload and application by selecting from a broad family of systems built from our flexible and reusable building blocks that support a comprehensive set of form factors, processors, memory, GPUs, storage, networking, power, and cooling solutions (air-conditioned, free air cooling or liquid cooling).  

Supermicro, Server Building Block Solutions, and We Keep IT Green are trademarks and/or registered trademarks of Super Micro Computer, Inc.

All other brands, names, and trademarks are the property of their respective owners.

 

View original content to download multimedia:https://www.prnewswire.com/news-releases/supermicro-begins-volume-shipments-of-nvidia-blackwell-ultra-systems-and-rack-plug-and-play-data-center-scale-solutions-302554330.html

SOURCE Super Micro Computer, Inc.

FAQ

What performance improvements do Supermicro's NVIDIA Blackwell Ultra systems offer?

The systems deliver 7.5x performance gains over NVIDIA Hopper systems, with 144 petaFLOPS of FP4 compute and 270 GB of HBM3e memory per GPU.

How much can customers save with Supermicro's DCBBS implementation?

Customers can achieve 40% power savings, 60% reduced data center footprint, 40% decreased water consumption, and 20% lower TCO.

What is the power consumption of Supermicro's new NVIDIA Blackwell Ultra GPUs?

The Blackwell Ultra's GB300 and B300 can be configured to use up to 1400W per GPU while offering 50% greater inferencing performance.

What networking capabilities do Supermicro's NVIDIA Blackwell Ultra solutions offer?

The solutions feature NVIDIA Quantum-X800 InfiniBand or Spectrum-X Ethernet compute fabrics with bandwidth up to 800 Gb/s.

What cooling options are available for Supermicro's NVIDIA Blackwell Ultra systems?

Supermicro offers both advanced air cooling and direct liquid cooling (DLC) technology options, with configurations available in 8U air-cooled and 4U liquid-cooled formats.