Supermicro (NASDAQ: SMCI) announced Thursday it has begun volume shipments of NVIDIA Blackwell Ultra systems, marking a major milestone in the rollout of next-generation AI infrastructure. The systems include NVIDIA HGX B300 platforms and GB300 NVL72 rack-scale solutions, aimed at powering global data centres, cloud providers, and enterprise AI operations.

A Big Bet on Blackwell Ultra

The new lineup features over 10 product configurations leveraging NVIDIA’s Blackwell and Blackwell Ultra architectures. Supermicro emphasized that its solutions are pre-validated, plug-and-play, available at the system, rack, and data center scale — tailored for rapid AI deployment.

Key specifications include:

  • Up to 1.1 exaFLOPS of FP4 compute with GB300 NVL72 racks.
  • 7.5x performance boost over Hopper-based systems with HGX B300 GPUs.
  • 1400W GPU power capacity, 50% more HBM3e memory, and 50% higher inference performance.
  • Cooling innovations: air or direct liquid cooling, with claims of 40% lower power use, 60% smaller footprint, and 40% less water consumption.

Supermicro has the best track record of fast and successful deployments of new NVIDIA technologies,” said Charles Liang, the company’s CEO, underscoring its strategy of moving quickly with cutting-edge hardware.

Financial & Strategic Context

With a market cap of $26.2 billion and 46.6% revenue growth over the past year, Supermicro is leaning on its strong liquidity (current ratio 5.25) to fuel this aggressive rollout. Analysts at Needham recently lifted their price target on Supermicro to $60 (from $39), maintaining a Buy rating despite weaker Q4 earnings tied to capital constraints.

The Blackwell Ultra launch comes as rivals like Lambda expand AI infrastructure using Supermicro servers, reflecting surging demand for enterprise-grade compute.

Broader AI Infrastructure Push

Supermicro’s systems support NVIDIA Quantum-X800 InfiniBand and Spectrum-X Ethernet fabrics, enabling up to 800 Gb/s bandwidth, critical for large-scale training and inference workloads.

The company’s push highlights intensifying competition in AI infrastructure, where efficiency, performance, and sustainability are becoming as important as raw power.

Disclosure: This article does not represent investment advice. The content and materials featured on this page are for educational purposes only.

Gold Surpasses Inflation-Adjusted Record High Set in 1980

CPI report shows prices climbing faster than July, What it means for Fed

France’s Government Collapses — What Moved in Markets and What’s Next

The Big Question: Are Crypto ETFs About to Explode?

ETF Boom or Bubble? US Now Has More ETFs Than Stocks as Retail Piles In

Bitcoin ETFs Surge on Trump Election Prospects, Market Braces for Volatility