Supermicro Collaborates with Arm to Deliver a New Class of Energy-Efficient Rack-Scale Infrastructure for Enterprise Agentic AI

Rhea-AI Summary

Supermicro (NASDAQ: SMCI) announced new rack-scale agentic AI infrastructure using Arm AGI CPUs, targeting higher performance-per-watt and data center efficiency.

Solutions include air- and liquid-cooled platforms that, according to Arm estimates, can deliver over 2x performance per rack and up to $10 billion CAPEX savings per gigawatt of AI capacity.

Positive

• Arm AGI CPU in Supermicro systems targets over 2x performance per rack versus traditional architectures
• Arm estimates up to $10 billion CAPEX savings per gigawatt of AI data center capacity
• Single air-cooled rack supports over 6,000 Arm AGI CPU cores
• OCP ORW rack design offers 45,696 cores and 336 Arm AGI CPUs per 48U rack
• Servers support up to 6TB DDR5-8800 RDIMMs per node for memory-intensive AI workloads

Negative

• None.

News Market Reaction – SMCI

+7.02%

64 alerts

+7.02% Session close to close

+4.0% Peak Tracked

-10.0% Trough Tracked

$30.85B Market Cap

0.7x Rel. Volume

In the Jun 2 session, SMCI gained 7.02%, reflecting a notable positive market reaction. Argus tracked a peak move of +4.0% during that session. Argus tracked a trough of -10.0% from its starting point during tracking. Our momentum scanner triggered 64 alerts that day, indicating high trading interest and price volatility.

Data tracked by StockTitan Argus on the day of publication.

Market Context

The stock moved +7.0% in the session following this news. A strong positive reaction aligns with SMC...

Analysis

The stock moved +7.0% in the session following this news. A strong positive reaction aligns with SMCI’s pattern of benefiting from AI-focused collaborations. Prior partnership/AI news produced modest average moves of 0.17%, so a large upside move on this Arm agentic AI rack-scale launch would highlight growing investor focus on high-density, performance-per-watt infrastructure and could be sensitive to broader AI hardware sentiment and follow-on execution.

Key Figures

Performance per rack: over 2x CAPEX savings: $10 billion per Gigawatt CPU cores: 136 cores +5 more

8 metrics

Performance per rack over 2x Arm AGI CPU vs traditional architectures in SMCI solutions

CAPEX savings $10 billion per Gigawatt Potential enterprise CAPEX savings for AI data center capacity (Arm estimate)

CPU cores 136 cores Arm AGI CPU dense microarchitecture

Memory bandwidth 6GB/s per core Arm AGI CPU memory bandwidth specification

Cores per air-cooled rack over 6,000 cores Agentic AI deployments in a single air-cooled rack

Memory per server up to 6TB DDR5-8800 RDIMMs Arm AGI CPU server configurations

GPU capacity (5U server) up to 8 Double Width GPUs 5U GPU server for AI training and inference

Cores per ORW rack 45,696 cores per rack 1U 4N configuration in a 48U ORW rack

Previous Partnership,AI Reports

2 past events · Latest: Oct 28 (Positive)

Same Type Pattern 2 events

Date	Event	Sentiment	24h Move	Catalyst
Oct 28	Government AI collaboration	Positive	+1.5%	Expanded NVIDIA collaboration and U.S.-based AI manufacturing for government uses.
Oct 09	Enterprise AI partnership	Positive	-1.2%	Strategic AI and storage alliance with Hitachi Vantara for enterprise workloads.

24h Move is the share-price change in the day after each event; other market factors may also have contributed.

Pattern Detected

Past partnership/AI announcements have produced modest, mixed price reactions, with one positive alignment and one divergence.

Recent Company History

Recent SMCI news highlights a consistent focus on large-scale AI infrastructure and strategic collaborations. Prior items include expanded NVIDIA-focused government AI solutions and a broad enterprise AI partnership with Hitachi Vantara, both tagged as partnership,AI and yielding an average move near flat. Alongside other recent AI factory and cloud deployments, today’s Arm-based agentic AI rack-scale launch fits an ongoing strategy of diversifying high-density, energy-efficient AI platforms.

Key Terms

agentic AI, rack-scale infrastructure, performance-per-watt, DDR5-8800, +3 more

7 terms

agentic AI technical

"The increasing compute demands of modern agentic AI require a new class of rack-scale infrastructure"

Agentic AI refers to computer systems that can make their own decisions and take actions without needing someone to tell them what to do each time. It's like giving a robot a degree of independence to solve problems or achieve goals on its own, which matters because it could change how we work and interact with technology in everyday life.

rack-scale infrastructure technical

"require a new class of rack-scale infrastructure that maximizes compute performance"

Rack-scale infrastructure is a way of building data center systems where servers, storage, and networking are grouped and managed as a single, modular unit at the level of a server rack rather than as many separate boxes. For investors, it matters because this design can lower capital and operating costs, speed upgrades, and improve reliability—similar to buying a furnished, move-in-ready apartment instead of assembling individual pieces of furniture, making scale and efficiency easier to predict.

performance-per-watt technical

"enables scalable, flexible infrastructure that maximizes performance-per-watt and dramatically lowers energy demand"

Performance-per-watt measures how much useful work a device or system can deliver for each unit of electrical power it consumes, essentially the speed or output achieved relative to energy used. Investors care because higher performance-per-watt means lower operating costs, longer battery life, smaller cooling needs and better competitiveness for products and data centers, which can translate into higher margins and stronger market adoption.

DDR5-8800 technical

"Up to 6TB of DDR5-8800 MT/s RDIMMs"

DDR5-8800 is a specification for a generation of computer memory (DDR5) that indicates a peak data transfer speed of 8,800 megatransfers per second. Think of it like a wider, faster highway for moving information inside computers: higher numbers mean quicker access to data and smoother performance for demanding tasks. For investors, widespread adoption or supply changes of this faster memory can influence sales, pricing and profit margins for memory makers and companies that rely on high-performance computing.

RDIMMs technical

"Up to 6TB of DDR5-8800 MT/s RDIMMs"

Registered DIMMs (RDIMMs) are a type of server-grade memory module that include a small built-in register which buffers control signals between the memory chips and the computer’s memory controller. Like adding a traffic coordinator on a busy road, the register reduces electrical strain and improves stability, allowing systems to support larger amounts of RAM and run more reliably; that matters to investors because RDIMM use affects server cost, performance, upgradeability and supplier demand in data-center and enterprise markets.

liquid-cooled technical

"as well as a liquid-cooled multi-node solution designed specifically for rack-scale agentic AI deployments"

A liquid-cooled system uses a circulating fluid (like water or a special coolant) to carry heat away from electronic components or machinery instead of relying on air. For investors, this signals equipment designed for higher performance, tighter temperature control and often greater reliability, but also typically higher upfront cost and potential maintenance needs — much like a car’s radiator versus a basic fan.

GPU-optimized technical

"5U GPU-optimized rackmount servers, as well as a liquid-cooled multi-node solution"

Software, models, or systems described as “GPU-optimized” are specifically designed or tuned to run most efficiently on graphics processing units (GPUs), which are specialized chips that handle many tasks at once—think of them as a multi-lane highway instead of a single-lane road. For investors, GPU optimization matters because it can greatly boost processing speed, reduce operating cost per unit of work, and enable handling of larger or more complex workloads (like AI, simulation, or high-frequency data), which can translate into competitive advantage, faster product development, and clearer scalability of operations.

AI-generated analysis. How Rhea-AI works. Not financial advice.

See more from StockTitan in Google Search and AI answers. Adds StockTitan as a preferred source · opens Google

Add on Google Not now

• Arm AGI CPU is designed for maximum efficiency in agentic AI workload orchestration
• Supermicro rack-scale Arm AGI CPU designs increase agentic AI compute performance density within any power envelope
• Supermicro's DCBBS capabilities and global manufacturing footprint reduce time-to-online (TTO) for large-scale AI infrastructure deployments

SAN JOSE, Calif. and TAIPEI, June 2, 2026 /PRNewswire/ -- Super Micro Computer, Inc. (NASDAQ: SMCI), an AI, Enterprise, Storage, 5G/Edge total solution provider featuring Data Center Building Block Solutions^® (DCBBS), today announced a new class of AI-centric solutions featuring Arm^® AGI CPUs. The increasing compute demands of modern agentic AI require a new class of rack-scale infrastructure that maximizes compute performance within the power envelopes and physical footprints of enterprise data centers. Supermicro's new solutions are built to support the rapid growth of agentic AI, delivering performance, efficiency, and density that maximizes the economics of rack-scale deployments backed by Supermicro's, end-to-end DCBBS capabilities reduce time-to-online.

"Supermicro continues to lead the industry when it comes to deploying new and innovative rack-scale solutions that maximize performance and efficiency," said Charles Liang, president and CEO of Supermicro. "Our DCBBS technology stack delivers end-to-end data center solutions of any size, which combined with the new density and efficient performance optimized Arm AGI CPU microarchitecture, helps enterprises realize significant TCO savings on their agentic AI infrastructure investments."

Learn more about the Supermicro Arm server line here.

"Agentic AI is driving a fundamental shift in infrastructure requirements, where efficiency, scalability, and orchestration performance are becoming just as critical as raw compute," said Mohamed Awad, Executive Vice President, Cloud AI Business Unit, Arm. "By combining Arm AGI CPUs with Supermicro's rack-scale system expertise, we're enabling infrastructure designed to deliver higher AI throughput, maximum compute density, and improved data center economics at scale."

Supermicro's new computing platforms consist of air-cooled dual-socket 2U compute-optimized and 5U GPU-optimized rackmount servers, as well as a liquid-cooled multi-node solution designed specifically for rack-scale agentic AI deployments. Combining Supermicro's proven modular, high-density architectures with the energy-efficient Arm Neoverse^® CSS V3-based CPUs enables scalable, flexible infrastructure that maximizes performance-per-watt and dramatically lowers energy demand to accelerate AI adoption across modern data centers.

When deployed in Supermicro solutions, the Arm AGI CPU can deliver over 2x performance per rack compared to traditional architectures and help enterprises save up to $10 billion in CAPEX per Gigawatt of AI data center capacity based on Arm's estimates. Building on Supermicro's industry-leading rack density and performance-per-watt, these solutions help ensure maximum utilization of data center space and power resources.

Arm AGI CPU boasts a dense 136-core microarchitecture purpose built for performance, minimizing legacy overhead and completing more work per cycle for sustained, unthrottled performance. 6GB/s memory bandwidth per core and latency-optimized memory access support linear scaling, while expanded memory capacity and flexible I/O provides energy-efficient, scalable agentic AI infrastructure to orchestrate thousands of parallel tasks across distributed infrastructure.

With over 6,000 cores in a single air-cooled rack, enterprises can efficiently deploy numerous dedicated systems for a high volume of agentic AI tasks.

The Supermicro lineup of Arm-based servers includes five models:

2U Hyper Server – Optimized for agentic AI, Cloud, and memory-intensive workloads
https://www.supermicro.com/en/products/system/hyper/2u/ars-222h-nr

• Two Arm AGI CPUs, up to 136 cores per CPU
• Up to 6TB of DDR5-8800 MT/s RDIMMs
• Up to two GPUs

5U GPU Server – GPU-dense configuration for AI training and inference
https://www.supermicro.com/en/products/system/gpu/5u/ars-522gp-nr

• Two Arm AGI CPUs, up to 136 cores per CPU
• Up to 6TB of DDR5-8800 MT/s RDIMMs
• Up to 8 Double Width GPUs

2U4N Liquid-Cooled Server – For OCP ORV3 environment
https://www.supermicro.com/en/products/system/hyper/2-ou/ars-242tp-qnr-lcc

• Two Arm AGI CPUs per node, up to 136 cores per CPU
• 4 Nodes in 2-OU potentially up to 20,672 cores per one ORV3 Rack.
• Up to 6TB of DDR5-8800 MT/s RDIMMs per node

2U Hyper-E Server – Single-socket edge-optimized architecture with front I/O
https://www.supermicro.com/en/products/system/hyper/2u/ars-212he-fnr

• Single socket Arm AGI CPU, up to 136 cores
• Up to 3TB of DDR5-8800 MT/s RDIMMs
• Up to 2 GPUs

1U 4N in an OCP ORW rack – Massive Compute Density

• ORW - 48U rack
• 336 Arm AGI CPUs per rack
• 168 Servers per rack with 45,696 cores per rack

Supermicro continues to lead the industry with its comprehensive portfolio of AI infrastructure solutions, enabling organizations worldwide to deploy scalable, efficient, and environmentally responsible AI data centers.

The latest rack-scale solutions will be on display at the Supermicro booth at Taipei Nangang Exhibition Center Hall 1, 4F, N0602, offering attendees a first-hand look at their design and capabilities.

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-collaborates-with-arm-to-deliver-a-new-class-of-energy-efficient-rack-scale-infrastructure-for-enterprise-agentic-ai-302786260.html

SOURCE Super Micro Computer, Inc.

FAQ

What did Supermicro (NASDAQ: SMCI) announce about Arm AGI CPUs for agentic AI on June 2, 2026?

Supermicro announced new rack-scale AI solutions built around Arm AGI CPUs to improve performance-per-watt and density. According to Supermicro, these platforms target modern agentic AI workloads with optimized efficiency, scalability, and orchestration across air- and liquid-cooled server designs.

How much performance per rack can Supermicro Arm AGI CPU systems deliver for AI workloads?

Supermicro systems with Arm AGI CPUs can deliver over 2x performance per rack compared with traditional architectures. According to Supermicro, this higher throughput is designed to maximize compute density and data center utilization for large-scale agentic AI deployments.

What potential CAPEX savings are associated with Supermicro and Arm AGI CPU AI data centers?

Arm estimates Supermicro solutions using Arm AGI CPUs may help save up to $10 billion in CAPEX per gigawatt of AI capacity. According to Supermicro, improved performance-per-watt and rack density are intended to enhance AI data center economics at scale.

Which Supermicro server models use Arm AGI CPUs for enterprise agentic AI workloads?

Supermicro introduced five Arm-based servers, including 2U Hyper, 5U GPU, 2U4N liquid-cooled, 2U Hyper-E, and 1U 4N ORW rack systems. According to Supermicro, these platforms target agentic AI, cloud, edge, and GPU-intensive applications.

How many CPU cores can Supermicro Arm AGI CPU racks provide for AI infrastructure?

Supermicro designs support over 6,000 cores in a single air-cooled rack and 45,696 cores in an ORW rack. According to Supermicro, this density enables many dedicated systems for high-volume agentic AI task orchestration.

What are the key technical features of the Arm AGI CPU used in Supermicro AI servers?

Arm AGI CPU features a dense 136-core microarchitecture with 6GB/s memory bandwidth per core and latency-optimized access. According to Supermicro, these specifications support linear scaling, high throughput, and energy-efficient agentic AI workload orchestration.

Where will Supermicro showcase its new Arm AGI CPU rack-scale AI solutions?

Supermicro will display its latest Arm-based rack-scale AI platforms at the Taipei Nangang Exhibition Center Hall 1, 4F, N0602. According to Supermicro, attendees can see the systems’ design and capabilities for enterprise agentic AI deployments firsthand.