Supermicro Delivers Performance and Efficiency Optimized Liquid-Cooled and Air-Cooled AI Solutions with AMD Instinct™ MI350 Series GPUs and Platforms

Rhea-AI Summary

Supermicro (SMCI) announces new H14 GPU solutions featuring AMD Instinct MI350 Series GPUs, offering both liquid-cooled and air-cooled options. The systems combine dual AMD EPYC 9005 CPUs with AMD Instinct MI350 GPUs, delivering 40% more tokens-per-dollar compared to competitors. Key features include 2.304TB total HBM3e memory per 8-GPU server, 288GB HBM3e per GPU (1.5x increase), 8TB/s bandwidth, and 1.8x petaflops of FP16/FP8 performance. The liquid-cooled 4U system with AMD Instinct MI355X GPUs reduces power consumption by up to 40%. The solutions are designed for AI training, inference, and HPC workloads, featuring AMD's 4th Gen CDNA architecture and new FP6/FP4 data types for enhanced AI capabilities.

Positive

• 40% more tokens-per-dollar efficiency compared to competitors
• Up to 40% power consumption reduction with new Direct Liquid Cooling architecture
• 1.5x memory capacity increase with 288GB HBM3e per GPU
• 1.8x performance improvement in FP16/FP8 compared to previous generation

Negative

• None.

Insights

Technology Infrastructure Analyst

Supermicro expands AI portfolio with AMD's newest GPUs, strengthening competitive position in high-performance computing market.

Supermicro's announcement of new liquid-cooled and air-cooled GPU solutions featuring AMD's latest Instinct MI350 series GPUs represents a significant expansion of their AI infrastructure portfolio. These H14 generation systems pair AMD EPYC 9005 CPUs with the new GPUs to create solutions optimized for large-scale AI workloads.

The technical specifications are impressive, with each 8-GPU server offering 2.304TB total HBM3e memory – a 1.5x increase over previous generations – and 1.8x petaflops of FP16/FP8 performance compared to the previous Instinct MI325X. This substantial memory capacity (288GB HBM3e per GPU) with 8TB/s bandwidth is crucial for handling increasingly large AI models efficiently.

What's particularly notable is Supermicro's enhanced Direct Liquid Cooling architecture, which can reduce power consumption by up to 40%. This addresses a critical challenge in AI infrastructure – managing the enormous power and cooling requirements of dense GPU deployments. By offering both liquid-cooled (4U) and air-cooled (8U) options, Supermicro is providing flexibility for different deployment environments while maximizing computational density.

The new systems leverage the open OCP Accelerator Module standard, maintaining Supermicro's commitment to industry-standard architectures. According to AMD CEO Lisa Su, these GPUs deliver 40% more tokens-per-dollar compared to competitors, suggesting a compelling value proposition for companies deploying large language models.

These systems are particularly well-positioned for the expanding AI inference market, where companies need to deploy trained models efficiently at scale. The support for new FP6 and FP4 data types further optimizes these systems for AI inference workloads, potentially opening new market opportunities for Supermicro beyond traditional high-performance computing customers.

• Supermicro launches highly optimized AI solutions, based on AMD Instinct MI350 Series GPUs and AMD ROCm™ Software, delivering breakthrough inference performance and power efficiency.
• The new Supermicro H14 GPU solutions are powered by the newest 4^th generation AMD CDNA™ architecture, delivering optimized performance and efficiency for large AI training models and high-speed inference workloads
• Large memory capacity with 2.304TB total HBM3e per 8-GPU server which delivers faster computation and more efficient scaling for AI, inferencing, and training

SAN JOSE, Calif., June 12, 2025 /PRNewswire/ -- Supermicro, Inc. (NASDAQ: SMCI), a Total IT Solution Provider for AI, Cloud, Storage, and 5G/Edge, is announcing that both liquid-cooled and air-cooled GPU solutions will be available with the new AMD Instinct MI350 series GPUs, optimized for unparalleled performance, maximum scalability, and efficiency. The Supermicro H14 generation of GPU optimized solutions featuring dual AMD EPYC™ 9005 CPUs along with the AMD Instinct MI350 series GPUs, are designed for organizations seeking maximum performance at scale, while reducing the total cost of ownership for their AI-driven data centers.

"Supermicro continues to lead the industry with the most experience in delivering high-performance systems designed for AI and HPC applications," said Charles Liang, president and CEO of Supermicro. "Our Data Center Building Block Solutions^® enable us to quickly deploy end-to-end data center solutions to market, bringing the latest technologies for the most demanding applications. The addition of the new AMD Instinct MI350 series GPUs to our GPU server lineup strengthens and expands our industry-leading AI solutions and gives customers greater choice and better performance as they design and build the next generation of data centers."

Learn more about Supermicro Servers with AMD Instinct MI350 Series GPUs here.

Supermicro's H14 generation data center solutions offer unparalleled performance and flexibility for a wide range of AI, HPC, Cloud, and Enterprise workloads. These systems are built on Supermicro's proven building-block architecture to help enterprise customers efficiently upgrade and scale their workloads. The new Supermicro GPU servers feature both the latest AMD EPYC CPU and AMD Instinct GPUs which together offer a powerhouse family of AI-enabled solutions, adding to Supermicro's data center building block solutions (DCBBS) ensuring customer success from consultation to on-site deployment and support.

"Our new AMD Instinct MI350 series delivers up to 40 percent more tokens-per-dollar relative to the competition, while retaining the industry standard form factor for OEM infrastructure compatibility," said Dr. Lisa Su, CEO and Chair, AMD. "By combining these GPUs with Supermicro's proven platforms, their customers can deploy fully integrated, air- or liquid-cooled racks built entirely on AMD technology, giving them the flexibility and efficiency they need to deploy robust AI solutions at any scale."

Supermicro is expanding its proven liquid-cooled and air-cooled high-performance fabric 8-GPU system, leveraging the open multi-architecture industry standard OCP Accelerator Module (OAM) to support the latest AMD Instinct MI350 series GPUs. For higher-density environments, the 4U liquid-cooled system with AMD Instinct MI355X GPUs features Supermicro's new improved Direct Liquid Cooling (DLC) architecture incorporating new technologies for cooling various server components and reducing power consumption by up to 40%, unlocking higher performance per rack and supporting advanced cooling infrastructure at scale. Customers can choose Supermicro's 4U liquid-cooled option for higher- density rack-scale deployments as well as the 8U option for air-cooled environments.

Designed to maximize computational throughput, memory bandwidth utilization, and energy efficiency to enable more power-efficient AI inference. these accelerated GPU servers with AMD Instinct MI350 series will offer 288GB HBM3e per GPU, boosting 1.5x memory capacity compared to previous generations of AMD Instinct accelerator, 8TB/s bandwidth, and 1.8x petaflops of FP16 / FP8 compared to the previous generation Instinct MI325X, so customers can process more data faster for their AI workloads.

"AI models aren't just increasing in size; they're demanding faster, more efficient infrastructure that can stay consistently deployable in real-world environments," said Paul Schell, Industry Analyst at ABI Research. "By adding AMD MI350 series GPUs into its latest H14 generation, Supermicro demonstrates a commitment to providing scalable, high-performance and memory solutions that are optimized for both AI training and inference. With support for both liquid and air cooling, these systems offer the flexibility and efficiency that CSPs, neo-clouds and enterprises alike need to power the next wave of AI growth."

These new GPU solutions are designed for powering AI at scale, across cloud service providers and enterprise partners, built on the 4th Gen AMD CDNA architecture, the AMD Instinct MI350 Series delivers exceptional performance and energy efficiency for a wide range of workloads, from AI training and inference to complex scientific simulations. The new AMD MI350 series will also feature new FP6 and FP4 data types to provide exceptional AI capabilities, handling larger models for AI deployments.

These Supermicro GPU systems supporting the AMD Instinct GPUs are available to order:

Liquid-Cooled GPU, 4U https://www.supermicro.com/en/products/system/gpu/4u/as%20-4126gs-nmr-lcc

Air-Cooled GPU, 8U
https://www.supermicro.com/en/products/system/gpu/8u/as%20-8126gs-tnmr 

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 manufacturer 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-delivers-performance-and-efficiency-optimized-liquid-cooled-and-air-cooled-ai-solutions-with-amd-instinct-mi350-series-gpus-and-platforms-302480186.html

SOURCE Super Micro Computer, Inc.

FAQ

What are the key features of Supermicro's new AMD Instinct MI350 GPU solutions?

The solutions feature dual AMD EPYC 9005 CPUs, 2.304TB total HBM3e memory per 8-GPU server, 288GB HBM3e per GPU, 8TB/s bandwidth, and 1.8x petaflops of FP16/FP8 performance

How much power reduction does SMCI's liquid-cooled system achieve?

The 4U liquid-cooled system with AMD Instinct MI355X GPUs reduces power consumption by up to 40%

What performance improvement does the AMD Instinct MI350 series offer over competitors?

The AMD Instinct MI350 series delivers 40% more tokens-per-dollar relative to the competition

What cooling options are available for Supermicro's new GPU solutions?

Supermicro offers both a 4U liquid-cooled option for higher-density rack-scale deployments and an 8U option for air-cooled environments

What is the memory capacity improvement in the new AMD Instinct GPUs?

The new GPUs offer 1.5x memory capacity increase compared to previous generations, with 288GB HBM3e per GPU