Powering Microsoft’s Azure Cobalt 200 with Arm Neoverse CSS V3: The next generation of Arm-based compute for the AI era
In the era of artificial intelligence (AI), a clear shift has occurred from commodity off-the-shelf systems to purpose-built infrastructure. Every workload, from traditional web services to scalable data analytics and large-scale model inference now plays a role in the AI-enabled intelligence pipeline. Modern data centers must be architected not as collections of discrete compute types, but as converged systems capable of handling diverse and demanding workloads efficiently. The industry is optimizing compute from the ground up to meet this need, prioritizing performance, scalability, and energy efficiency.
Azure Cobalt 200: Accelerating the converged AI data center
Today, Microsoft announced Cobalt 200, the next generation of Arm-based CPU processors powering its Azure cloud infrastructure and the first publicly announced silicon built on the Arm Neoverse Compute Subsystem V3 (CSS V3). Cobalt 200 exemplifies Microsoft’s vision of the converged AI data center, where general-purpose CPUs and specialized accelerators work side by side with bespoke networking, storage, and security offload to train, fine-tune, and deploy cloud and AI workloads at a massive scale.
A partnership for the future of AI infrastructure
Arm and Microsoft share a vision for a more efficient, scalable, and sustainable cloud — one where every layer of the stack, from silicon to software, is optimized not just for token generation — but for complete end-to-end intelligence per watt. The launch of Cobalt 200 marks another milestone in this collaboration, demonstrating what’s possible when world-class computing architecture meets world-class cloud engineering.
“The shift to purpose-built systems for AI and cloud-native workloads is transforming how infrastructure is designed,” said Dermot O’Driscoll, VP Products and Solutions, Cloud AI Business Unit, Arm. “Our collaboration with Microsoft on Cobalt 200 demonstrates how Arm Neoverse CSS V3 can be leveraged to accelerate this transformation and enable highly optimized silicon for a wide range of data-driven applications.”
“Cobalt 200 is the result of our longstanding co-innovation efforts between Microsoft and Arm, demonstrating what is possible when optimizing silicon, server, and software together,” said Selim Bilgin, Corporate Vice President of Silicon Engineering, Microsoft. “We’re excited to be the first implementers of the Neoverse CSS V3, and taking advantage of Arm’s extensible architecture to implement unique capabilities like per-core DVFS, custom compression and cryptographic accelerators, as well as Azure Boost. Together, we are delivering compute infrastructure that improves performance and power efficiency for real cloud workloads.”
Cobalt 200 follows our successful collaboration with Microsoft on its Cobalt 100 processor, which is delivering Neoverse-powered instances that are outpacing traditional competitors on performance and efficiency across various real-world workloads, including database, networking, and quantitative finance.
Accelerating custom silicon innovation with Neoverse CSS
Neoverse CSS V3 is a pre-integrated and silicon-proven platform that enables silicon designers — from hyperscalers to Arm Total Design partners including ASIC design houses, silicon startups or established merchant silicon providers — to design and deploy custom silicon faster than ever before. It is built around Arm’s most performant Neoverse CPU core to date and includes support for Armv9 architectural features such as Confidential Computing Architecture (CCA) and Scalable Vector Extension 2 (SVE2).
This approach allows partners to focus on innovation where it matters most — optimizations in silicon and at the system level specific to their workload profile — while Arm continues to deliver the world’s leading platform for cloud-to-edge compute, backed by a cohesive cloud-to-edge AI-enabled software ecosystem. Microsoft did exactly this by modeling more than 350,000 configuration parameters based on extensive workload profiling to optimize performance for both customer needs and internal applications like Microsoft Teams.
Building the future cloud AI data center
As AI continues to permeate the data center conversation, compute must optimize for every workload — certainly AI training and inference, but also the critical web, storage and data serving and processing workloads that form the backbone of modern commerce and are critical in the AI era to translate tokens into intelligence. This macro-level convergence of general-purpose and AI-optimized compute has become the foundation of next-generation cloud infrastructure. With Cobalt 200 powered by Neoverse CSS V3, Arm and Microsoft are a fundamental part of this computing transformation in shaping more performant, efficient and flexible infrastructure for the future intelligence-optimized data center.
Any re-use permitted for informational and non-commercial or personal use only.
Editorial Contact
Latest on X
We’ve all heard the talk of an AI bubble, but with massive investment and real opportunities ahead, Rene Haas and @Arista_Networks CEO Jayshree Ullal dig into what makes this AI moment different, and how it can truly work and scale in practice: https://okt.to/2eGs54
We're so proud for our partnership with @UNHCRInnovation @UNRefugeesUK was shortlisted for the Best Tech Inclusion Initiative at the Tech Impact Awards.
A reminder that innovative technology can address some of the world’s most pressing humanitarian challenges. 💡🎉
Arm is delivering the performance, scale, and efficiency required to enable energy-efficient compute in more places. At #FortuneBrainstormAI CEO Rene Haas will will share his vision for the future of AI built on Arm. ⚡
📍 December, 8th @ 5:15 PM
https://okt.to/ZeXDBI
The vivo X300 series marks a major leap forward for mobile AI, powered by the Arm Lumex platform, delivering faster, smarter, and more efficient mobile experiences.
Eric Xia, Chief Chipset Planning Expert, shares how we brought it to life: https://okt.to/UpRN4v
Huge congratulations to our partners at @awscloud who have just announced Graviton5, the company’s most powerful and efficient CPU to date. 🎉
It's an exciting step towards the future of powerful, efficient, purpose-built compute.
AWS introduces Graviton5—the company’s most powerful and efficient CPU
Fifth generation chip provides the best price performance for a broad range of workloads in Amazon EC2.
okt.to
Purpose-built compute is fueling the AI data center, and @awscloud Graviton shows why. As the host CPU for Trainium in AWS AI UltraServers, Arm delivers the performance, power, and scale hyperscalers depend on.
The converged AI data center is here. 🚀
How Arm is redefining compute through the converged AI data center
Hyperscalers are standardizing on Arm to power the converged AI data center. Explore how AWS Graviton5 and Arm Neo...
okt.to
Purpose-built compute is fueling the AI data center, and @awscloud Graviton shows why. As the host CPU for Trainium in AWS AI UltraServers, Arm delivers the performance, power, and scale hyperscalers depend on.
The converged AI data center is here. 🚀 https://okt.to/0mzKJU
Achieve greater integration, intelligence, and control for OEMs? Greater flexibility, faster iteration, and less friction across product lifecycles for developers?
Arm is powering this shift with flexible, AI-ready, energy-efficient compute at the edge. 💪…
We were delighted to welcome H.R.H. Prince Daniel and the inspiring entrepreneurs from Prince Daniel’s Fellowship to Arm Cambridge!🙌
Great conversations, bright ideas, and lots of energy around the future of innovation.
Thanks for spending the day with us! 🤝
It started in karting. Now, @1jessicahawkins is chasing the toughest race in the world, Le Mans.
Fueled by determination and innovation, and supported by the @AstonMartinF1 team and Arm.
This is Jessica's story.
🎥 Watch now: https://shorturl.at/GfycI
5 cities. 1 vision. 👏
Arm Unlocked 2025 in Shanghai, Shenzhen, Seoul, Taipei, and Tokyo brought the global ecosystem together to define how AI scales, from edge and physical AI to cloud AI, all on the Arm compute platform.
https://okt.to/q3PXFU
What’s your favorite way to celebrate milestones at work?
35 years of innovation. 35 years of people shaping the future of computing. 💪
Our offices around the world celebrated 35 years of Arm in true style — connection, creativity, and (of course) cake 🎉
Today, we're celebrating 35 years of Arm! 🎉
That makes us old enough to remember dial-up, and young enough to be at the leading edge of AI.
Arm technology touches 100% of the connected global population, with more than 325 billion chips shipped worldwide.…
This week, we're celebrating 35 years of Arm! 🎉
That makes us old enough to remember dial-up, and young enough to be at the leading edge of AI.
Arm technology touches 100% of the connected global population, with more than 325 billion chips shipped worldwide.…
Mohamed Awad, EVP of Cloud AI Business Unit at Arm, recently joined Silicon Valley Tech Talks to discuss how AI is reshaping data centers, and how Arm’s Neoverse and Arm Total Design ecosystem are helping the industry meet the growing demands of compute.
Arm's Innovations for AI Infrastructure
...
okt.to
Collaboration drives impact. 📷
Over the past decade, we've partnered with @UNICEFinnovate
to apply technology for good, developing scalable AI solutions to protect children’s health, advance education, and build climate resilience.
Discover the story @FT:…
#MSIgnite may be over but if you missed us there's no need to worry.
Head over to our Cloud Migration Hub or Developer Program resources to learn more about the future of cloud computing on Arm. 💪 https://okt.to/gTJ8qj
Inspired by #SC25? So are we.
From record-breaking Arm-based supercomputers to powerful new innovations in AI and HPC, the future of compute is being built on Arm.
Be part of what’s next 👉 http://careers.arm.com
From building networks to shaping the AI era, @AristaNetworks’ Jayshree Ullal joins Rene Haas on the Tech Unheard podcast to talk bold moves, software-first thinking, and leading through change in the AI era.
Listen to the new episode now: https://okt.to/9dEpyc 🎧
Micosoft Azure Cobalt 100 VM powered by Arm Neoverse deliver up to 99% better price-performance across workloads. From web infrastructure to quantitative finance, these instances are enabling efficiency, scalability, and real-world value .
Built on Arm - redefining the future.…
📢 Introducing Cobalt 200!
In partnership with @Microsoft we're bringing you the first publicly announced silicon built on the Arm Neoverse Compute Subsystem V3 (CSS V3).
A vital part of our commitment to a more efficient, scalable, and sustainable cloud!…
👋 Good morning from #SC25!
Stop by the Arm booth to explore our latest demos, connect with our talent team, and learn about open roles and life at Arm.
💡 Discover how you can help shape the future of AI and HPC — and be part of the team driving the future of compute! 📍#4425
🚀 We kicked off #SC25 with @AWSCloud & @NVIDIA, bringing the Arm HPC & Advanced Compute community together to connect and share their experiences building on Arm.
With every major hyperscaler choosing Arm, we’re powering the future of AI and supercomputing. See you at the show!
