Arm Physical IP to Accelerate Mainstream Mobile and IoT SoC Designs on TSMC 22nm ULP/ULL Platform
SAN JOSE, Calif. – Arm announced today its Arm® Artisan® physical IP will be used in TSMC’s 22nm ultra-low power (ULP) and ultra-low leakage (ULL) platform for Arm-based SoCs. TSMC 22nm ULP/ULL is optimized for mainstream mobile and IoT devices, enabling improved performance for Arm-based SoCs, and reductions in both power and area when compared to the previous-generation TSMC 28nm HPC+ platform.
“This next-generation process technology addresses more functionality using less power and at a smaller area size,” said Gus Yeung, general manager, physical design group, Arm. “The combination of Artisan Physical IP and the design and manufacturing cost advantages of TSMC’s 22nm ULP/ULL platform will provide our mutual partners with immediate compute performance-per-milliwatt and area benefits.”
Artisan physical IP for TSMC 22nm ULP and ULL process technologies includes foundry-sponsored memory compilers optimized for the low-leakage and low-power requirements of next-generation edge computing devices. Complementary to those compilers are ultra-high-density and high-performance physical IP standard cell libraries including power-management kits, and thick-gate oxide cells to help optimize low-leakage power. General Purpose I/O solutions are also offered to ensure maximum performance, power and area (PPA) optimization.
“Artisan Physical IP enables TSMC to accelerate tape-out timelines and bring leading SoCs to market faster for mainstream IoT and mobile devices,” said Suk Lee, TSMC senior director, Design Infrastructure Marketing Division. “Building on our successful 28nm HPC+ platform collaboration, TSMC and Arm are delivering significant power and area reductions, giving our mutual silicon partners opportunities to enable an improved edge compute experience across more devices.”
Arm’s physical IP is a trusted, widely used solution with more than 10 billion integrated circuits (ICs) shipped by Arm partners every year. The aggressive integration schedule of TSMC 22nm ULP and ULL process technologies is on-track for Arm and TSMC silicon partners targeting 2H 2018 tape-outs.
Any re-use permitted for informational and non-commercial or personal use only.
Media Contacts
Latest on X
Stable Audio Open Small - built on Arm CPUs by @StabilityAI - is now live! 🚀
What’s new?
➡️ Step-by-step learning path for Android deployment
➡️ Open-source code you can copy, run, or drop into your own app
Build it. Remix it. Share it.
🔗 Get started: https://okt.to/YXyZvC
“The reality of security is, there isn’t one silver bullet for for all of it” 🔒
From securing AI models to defending against emerging threats across compute, hear Chief architect Richard Grisenthwaite unpack AI and hardware security with @sallywf.
Hardware Security in the Age of AI - EE Times Podcast
Explore AI's dual role as both a threat and a solution in hardware security with insights from Arm's chief architect!
okt.to
At #LinaroConnect this week?
Catch the keynote walk-ons to hear about the robust Arm software ecosystem from Cloud to Edge, and how collaborating with @LinaroOrg can help you accelerate innovation within the Arm ecosystem:
Connect | Linaro
Linaro Connect is a series of events that brings together the Arm Ecosystem. This is the ONLY place where develo...
okt.to
Today at #FTCar, Dipti Vachani joined @VolvoGroup’s Christopher Lindstrom, @BentleyMotors 's Phil Gush & @ftlive’s Tim Bradshaw to explore chip supply, the AI-defined era & what it takes to build a resilient automotive ecosystem—from SDVs to smarter supply strategies.
📸…
Curious about AI?
Learn the fundamentals of AI & ML, train real models, and explore ethical challenges in just 4 weeks.
Taught by Arm engineers + top academics.
Start your AI journey today 👉
The expansion of the PyTorch Foundation is an amazing opportunity to unlock further AI innovation from cloud to edge. Arm is proud to support this next chapter in open-source AI 🤝
PyTorch Foundation has expanded into an umbrella foundation. @vllm_project and @DeepSpeedAI have been accepted as hosted projects, advancing community-driven AI across the full lifecycle.
Supporting quotes provided by the following members: @AMD, @Arm, @AWS, @Google, @Huawei,…
For the past 10 years, Arm has supported @UNICEF in exploring how scalable, open source, and sustainable tech can help improve children’s health, education, climate resilience, and access to information. 🤝
Here’s what a decade of impact looks like 👇
We closed FYE25 with a new milestone - crossing $1B in quarterly revenue for the first time. 🎉
As AI continues to drive demand for Armv9, custom silicon, and Arm compute subsystems, we’re proud to enable the AI revolution at scale, from cloud to edge:
'@arcee_ai ran a 32 billion-parameter model on an Arm-based CPU. That’s not a typo.
On the latest Arm Viewpoints podcast, Arcee AI's Chief Evangelist @julsimon explains how quantization + clever engineering is paving the way for SLMs in the enterprise: https://okt.to/nrEkIK
'@Spotify. @GoogleCloud. Arm.
One (virtual) room. One big shift in cloud performance.
Hear how Arm Neoverse-based Google Axion processors are bringing better performance, efficiency, and TCO to Spotify's cloud-native and AI workloads. Sign up: https://okt.to/P3ronG
🗓️ Join us on Wednesday May 7 for our next financial results conference call.
After market close, we'll report our earnings results for the fourth quarter and fiscal year 2025, followed by a conference call to review our results and business outlook: https://okt.to/5JkpNx
⚡ If you've ever thought "5GHz must mean fast," think again.
@EposVox breaks down why the IPC (Instructions Per Cycle) metric for modern CPU performance best represents real-world user experiences and why it matters more than ever.
📺 Watch now.
Why your fast CPU still feels slow
This video is sponsored by Arm. Learn why IPC matters to mobile and how Arm is powering the future of smartpho...
okt.to
The latest tools from @AIatMeta - Llama Guard 4, Firewall and Prompt Guard 2 - help embed policy-grade safety into gen AI interactions.
With real-time multimodal filtering combined with Arm’s power-efficient compute platform, they support secure, scalable AI from cloud to edge.
Major updates from LlamaCon!
We’re advancing AI security with new open-source Llama protection tools and new AI- powered solutions for the defender community.
Developers can now access:
-- Llama Guard 4, a customizable safeguard that supports protections for text and image…
AI's future won’t be built on power alone; it will be built on smarter foundations.
Richard Grisenthwaite, Arm’s EVP and Chief Architect, shares why strong architectural foundations are key to scaling AI sustainably in @computing_news ⏬
Tackling AI’s challenges starts with strong architectural foundations
Tackling AI’s soaring energy demands needs both hardware, software and scientific advancements, says Arm EVP Richard Grisenthwaite
okt.to
From data centers to edge devices, CEO Rene Haas and @Synopsys' Sassine Ghazi are showcasing how collaboration is fueling the future of AI.
Find out what’s next in AI innovation from their conversation at the 35th annual Synopsys User Group: https://okt.to/NG8efa
Innovate alongside compliance. 🤝
In the AI Readiness Index, Sr. Director of Government Affairs Vince Jesaitis explores the evolving global regulatory landscape from the EU AI Act's risk-based approach to the U.S. sectoral model.
Download the report: https://okt.to/meYpHL
40 years ago today, the Arm architecture was born with five simple words:
"Hello World, I am ARM"
We were honored to welcome Sophie Wilson, Steve Furber and Jamie Urquhart back to our Cambridge HQ to celebrate the legacy they helped create and the future we’re building on Arm.
Celebrating 40 years of Arm architecture at the turkey barn where it all began. 👏
Earlier this week, we unveiled a new plaque in Swaffham Bulbeck honoring the architecture that sits at the foundation of modern computing, alongside the pioneers who helped shape it.
Chiplets are redefining the automotive industry. How do we fulfill their potential?
➕ Collaboration
➕ Standardization
➕ Resource sharing
We're joining forces with @imec_int's Automotive Chiplet Program to make it a reality. Find out how: https://okt.to/xAuDk1
Take a trip down memory lane with us. 🛣️
From the BBC Micro to AI-powered devices, the Arm architecture has shaped 40 years of innovation, powering smartphones, IoT, our cars, the cloud, and more.
Let’s look back at the tech that changed everything.
Deploying AI is just the start. The real challenge? Running it efficiently, securely and at scale.
In our report with @techreview we explore why heterogeneous compute is key to enabling edge AI everywhere from our phones to IIoT.
📥 See why: https://okt.to/lQVD69
The industry is developing comprehensive standards, like those from PSA Certified and others, to secure the next generation of custom silicon. 🔐
Learn more about how to secure the silicon solutions of the future in our recent report: https://okt.to/6ND1AB
As demand for performance and efficiency grows, tech giants are rethinking their infrastructure - and turning to Arm.
Mohamed Awad joins @EETimes to explore how our flexible, power-efficient architecture is shaping the future of computing:
Why Do Hyperscalers Design Their Own CPUs? - EE Times
With huge financial investment required to enter the realm of custom silicon design, why is this route so appealing to hyperscalers?
okt.to
Smaller models. Bigger impact. 💥
@Arcee_AI is using Arm-based CPUs to run SLMs in parallel for agentic AI. With 4x performance gains from quantization and KleidiAI, @julsimon shares why the future of AI is efficient, scalable - and built on Arm CPUs: https://okt.to/UtMRxA
