Arm Ethos-U85 NPU: Unlocking Generative AI at the Edge with Small Language Models
As artificial intelligence evolves, there is increasing excitement about executing AI workloads on embedded devices using small language models (SLM).
Arm’s recent demo, inspired by Microsoft’s “Tiny Stories” paper and Andrej Karpathy’s TinyLlama2 project, where a small language model trained on 21 million stories generates text, showcases endpoint AI’s potential for IoT and edge computing. In the demo, a user inputs a sentence, and the system generates an extended children’s story based on it.
Our demo featured Arm’s Ethos-U85 NPU (Neural Processing Unit) running a small language model on embedded hardware. While large language models (LLMs) are more widely known, there is growing interest in small language models due to their ability to deliver solid performance with significantly fewer resources and lower costs, making them easier and cheaper to train.
Implementing A Transformer-based Small Language Model on Embedded Hardware
Our demo showcased the Arm Ethos-U85 as a small, low-power platform capable of running generative AI, highlighting that small language models can perform well within narrow domains. Although TinyLlama2 models are simpler than the larger models from companies like Meta, they are ideal for showcasing the U85’s AI capabilities. This makes them a great fit for endpoint AI workloads.
Developing the demo involved significant modeling efforts, including the creation of a fully integer int8 (and int8x16) Tiny Llama2 model, which was converted to a fixed-shape TensorFlow Lite format suitable for the Ethos-U85’s constraints.
Our quantization approach has shown that fully integer language models can successfully balance the tradeoff between maintaining strong accuracy and output quality. By quantizing activation, normalization functions, and matrix multiplications, we eliminated the need for floating-point computations, which are more costly in terms of silicon area and energy—key concerns for constrained embedded devices.
The Ethos-U85 ran a language model on an FPGA platform at only 32 MHz, achieving text generation speeds of 7.5 to 8 tokens per second—matching human reading speed—while using just a quarter of its compute capacity. In a real system-on-chip (SoC), performance could be up to ten times faster, significantly enhancing speed and energy efficiency for AI processing at the edge.
The children’s story-generation feature used an open-source version of Llama2, running the demo on TFLite Micro with an Ethos-NPU back-end. Most of the inference logic was written in C++ at the application level. Adjusting the context window enhanced narrative coherence, ensuring smooth, AI-driven storytelling.
The team’s adaptation of the Llama2 model to run efficiently on the Ethos-U85 NPU required careful consideration of performance and accuracy due to the hardware limitations. Using mixed int8 and int16 quantization demonstrates the potential of fully integer models, encouraging the AI community to optimize generative models for edge devices and expand neural network accessibility on power-efficient platforms like the Ethos-U85.
Showcasing the Power of the Arm Ethos-U85
Scalable from 128 to 2048 MAC units (multiply-accumulate units), the Ethos-U85 achieves a 20% power efficiency improvement over its predecessor, the Ethos-U65. A standout feature of the Ethos-U85 is its native support for transformer networks, which earlier versions could not support.
The Ethos-U85 enables seamless migration for partners using previous Ethos-U NPUs, allowing them to capitalize on existing investments in Arm-based machine learning tools. Developers are increasingly adopting the Ethos-U85 for its power efficiency and high performance.
The Ethos-U85 can reach 4 TOPS (trillions of operations per second) with a 2048 MAC configuration in silicon. In the demo, however, a smaller configuration of 512 MACs on an FPGA was used to run the Tiny Llama2 small language model with 15 million parameters at just 32 MHz.
This capability highlights the potential for embedding AI directly into devices. The Ethos-U85 effectively handles such workloads even with limited memory (320 KB of SRAM for caching and 32 MB for storage), paving the way for small language models and other AI applications to thrive in deeply embedded systems.
Bringing Generative AI to Embedded Devices
Developers need better tools to navigate the complexities of AI at the edge, and Arm is addressing this with the Ethos-U85 and its support for transformer-based models. As edge AI becomes more prominent in embedded applications, the Ethos-U85 is enabling new use cases, from language models to advanced vision tasks.
The Ethos-U85 NPU delivers the performance and power efficiency required for innovative, cutting-edge solutions. Like the “Tiny Stories” paper, our demo represents a significant advancement in bringing generative AI to embedded devices, demonstrating the ease of deploying small language models on the Arm platform.
Arm is opening new possibilities for Edge AI across a wide range of applications, positioning the Ethos-U85 to power the next generation of intelligent, low-power devices.
Read how Arm is accelerating real-time processing for edge AI applications in IoT with ExecuTorch.
Any re-use permitted for informational and non-commercial or personal use only.
Editorial Contact
Latest on X
We've been recognized by @TIME & @StatistaCharts as one of the World's Best Companies in Sustainable Growth.
By providing the industry with the most power-efficient & highest-performing compute platform, we're paving the way to sustainable & secure AI: https://okt.to/UvMTc9
The way we work is just as important as what we do.
Paula shares a glimpse into what she values most in her role as Senior Principal Engineer & Team Lead.
Visit our careers site and learn how you can make an impact.
It was an incredible week at #SC24 for solutions built on the Arm compute platform.
#ICYMI: At this year's @HPCWire 2024 awards, Arm-based products using NVIDIA Grace swept the floor.
🎉 Congratulations to our partners - we're proud to build the future of computing together.
Las Vegas came alive this weekend with the Formula 1 Grand Prix. 🏎️🏁
Did you catch the Grand Prix? And the key question... did you spot our billboard? 👀
Arm Tech Symposia 2024 was bigger and better than ever!🚀
This year, we brought the technology ecosystem together across China, Japan, South Korea, and Taiwan to explore the huge opportunities AI offers today.
Explore the highlights in our latest blog: https://newsroom.arm.com/blog/arm-tech-symposia-2024?utm_source=twitter&utm_medium=social-organic&utm_content=blog&utm_campaign=mk19_events_apactechsymposia
Where is on-device AI headed next? 📱
In @talkandroid, James McNiven, VP of Product Management, shares details on what to expect from the next-generation of AI-driven smartphone experiences and how the Arm compute platform is enabling these. ⏬
Interview With Arm: What’s Next For On-device AI On Smartphones? - Talk Android
On-device AI on smartphones is going places, and we talked with Arm's James McNiven to find out what's next for this powerhouse technology
okt.to
With the momentum we've seen at #MSIgnite around Azure Cobalt 100 VMs, we're excited to work with the ecosystem to enable the AI-accelerated future in a sustainable, scalable way on Arm Neoverse.
Get up to speed on what you might have missed from the show https://newsroom.arm.com/blog/arm-and-microsoft-igniting-a-new-era-of-cloud-computing-for-ai?utm_source=twitter&utm_medium=social-organic&utm_content=blog&utm_campaign=mk03_infrastructure_na
As proud investors in Enfabrica following its Series C Round, we look forward to supporting the team’s incredible growth on Arm Neoverse, enabling scalable networking solutions that meet the needs of demanding AI workloads.
👏 Congratulations, @enfabrica_!
Perseverance is a fundamental part of who we are as a company.
In the new State of European Tech documentary from @atomico, Rene Haas shares how this has enabled us to become the industry's highest-performing and most power-efficient compute platform 🔽
The State of European Tech Anniversary Documentary
As The State of European Tech celebrates its tenth anniversary, the team has gathered the most influential voices ...
okt.to
This is what a day in the life looks like for Graduate Product Manager, Conall Farney. 📹
Since joining us as a Sales Intern, Conall moved into a new role matching his growing passions.
Discover your path with us: https://okt.to/KdWshB
Game-changing quantum breakthrough: Equal1 runs Arm Cortex at 3.3 Kelvin (-269.85°C)!
Equal1's innovative approach could make quantum computing as accessible as traditional servers.
From drug discovery to AI, the future is getting cooler ❄️🌡️ https://newsroom.arm.com/blog/pushing-quantum-computing-boundaries-with-arm-equal1?utm_source=twitter&utm_medium=social-organic&utm_content=blog&utm_campaign=mk13_tech-strat_na
Here at Arm, we know that prioritizing your wellbeing is key to unlocking your potential.
👉 That’s why we’re committed to nurturing our team’s passions to drive innovation. Scroll through to hear what some of our people have to say about how Arm helps them thrive.
When software architecture isn't standardized, hardware changes can cause issues.
@SchneiderUKI collaborated with us to solve this challenge by bringing in Arm SystemReady, where new hardware works with the software:
SystemReady Implementation at Schneider Electric: Accelerating Development and Enhancing Platform...
How Schneider Electric leveraged Arm SystemReady to accelerate development, enhance platform flexibility and overcome supp...
newsroom.arm.com
Game-changing quantum breakthrough: Equal1 runs Arm Cortex at 3.3 Kelvin (-269.85°C)!
Equal1's innovative approach could make quantum computing as accessible as traditional servers. From drug discovery to AI, the future is getting cooler 🌡️❄️ https://newsroom.arm.com/blog/pushing-quantum-computing-boundaries-with-arm-equal1?utm_source=twitter&utm_medium=social-organic&utm_content=blog&utm_campaign=mk13_tech-strat_na
AI is causing a fundamental shift in data center infrastructure.
With the flexibility of the Arm compute platform and our relentless focus on energy efficiency and customization, we're at the center of this transformation. Here's why 👇
Arm in the Data Center: The Dawn of a New Era - Creative Strategies
Introduction: 2024 - The Tipping Point for Arm's Data Center Momentum As we close out 2024, the semiconductor indu...
creativestrategies.com
The combined power and performance of Arm and @GitHub is optimizing the developer experience and unlocking advances in AI.
Here's to many more years of partnership and building the future of AI #onArm. 💪
From versatility and cost-effectiveness to scalability and power efficiency, here's 5 key reasons why CPUs are the top choice for AI ⏬
Top 5 Reasons why CPU is the Best Processor for AI Inference - Embedded
Advanced artificial intelligence (AI), like generative AI, is enhancing all our smart devices. However, a common misco...
www.embedded.com
From on-CPU AI processing to chiplet-enabled designs, hyperscalers get more control, efficiency, and power to meet the demands of AI when building #onArm.
@TheNextPlatform shares how custom Arm CPUs are transforming the data center.
Custom Arm CPUs Drive Many Different AI Approaches In The Datacenter
Sponsored Feature Arm is starting to fulfill its promise of transforming the nature of compute in the datacenter, ...
www.nextplatform.com
Arm founding CEO & former chairman, Sir Robin Saxby, has been inducted into the City of London Engineering Hall of Fame!
This acknowledges the transformational impact of Saxby's work on the Arm compute platform, laying the foundation for who we are today: https://newsroom.arm.com/blog/sir-robin-saxby-city-of-london-engineering-hall-of-fame?utm_source=twitter&utm_medium=social-organic&utm_content=blog&utm_campaign=mk29_exec-comms_na
We're entering a new partnership with @Panasonic Automotive Systems.
To accelerate development cycles in the automotive industry, we'll adopt and extend VirtIO, the device virtualization framework that will help standardize vehicle architecture for SDVs: https://newsroom.arm.com/news/arm-and-panasonic-partner-to-standardize-sdvs?utm_source=twitter&utm_medium=social-organic&utm_content=newsletter&utm_campaign=mk12_automotive_na
We're delighted to welcome Charlotte Eaton back to Arm. 🎉
Charlotte rejoins us as Chief People Officer, drawing on her wealth of experience at Arm and in other leadership roles to empower our people to do the best work of their careers: https://newsroom.arm.com/news/arm-announces-appointment-of-charlotte-eaton-as-chief-people-officer?utm_source=twitter&utm_medium=social-organic&utm_content=newsroom&utm_campaign=mk29_exec-comms_na
Demand for the Arm compute platform remains strong in the age of AI.
In Q2 FYE25, we achieved record royalty revenue for Q2, with strong YoY growth thanks to the increased demand for energy-efficient compute and more complex AI chips built #onArm:
Frederique Olivier relies on Arm to capture the beauty of the natural world. 🌎
With the help of Arm, she can shoot images with better quality and source power in extreme conditions.
Discover the world from Frederique's eyes, made possible #onArm: https://newsroom.arm.com/blog/purpose-on-arm-frederique-olivier-photographer?utm_source=twitter&utm_medium=social-organic&utm_content=blog&utm_campaign=mk08_brand_na
🗓️ Join us on Wednesday November 6 for our next financial results conference call.
After market close, we'll report our earnings for the second quarter of fiscal year 2025, followed by a conference call to review our results and business outlook: https://newsroom.arm.com/news/arm-announces-earnings-release-date-for-second-quarter-fiscal-year-ended-2025?utm_source=twitter&utm_medium=social-organic&utm_content=newsroom&utm_campaign=mk29_exec-comms_na
