Edge AI

Nordic Semiconductor use their experience in ultra-low-power to take wireless technology to the next level, allowing more powerful data processing at the edge, bringing significant benefits to your applications.

 Why use Edge AI on your Nordic SoC? 

icon      FEATURE-PROOF:  By using Edge AI on your device you can add new and innovative features to your product that makes it to stand out from the competitiors'
 Icon of two cogwheels BANDWIDTH & LATENCY: Process data locally in real-time - no need to use bandwidth to send raw data and no time wasted wating for a response from the cloud 
 icon POWER & EFFICIENCY:  Local processing uses less power than sending data over the air, allowing your device to have longer operation and smaller battery size

What is Neuton

The tiny-ML experts in Neuton have developed a unique neural network framework, not based on any existing technologies such as TensorFlow, PyTorch or similar, to create ultra-tiny ML models. This framework allows building models 10x smaller than competing frameworks, down to as small as single-digit kilobytes, with great generalizing capabilities. This gives the models the same or better accuracy and faster inferencing than these competing frameworks, resulting in lower power consumption. As of 2025, Neuton is now a part of Nordic Semiconductor.

We are currently working to integrate Neuton natively into our development ecosystem, adding the tools, firmware examples and support that ease the life of developers and add value to their applications. If you’re really eager to try out building and integrating a Neuton model yourself, you can test it out using their legacy online platform

What does it mean for developers?

Developers looking at using Edge AI in their application can read our DevZone blog, where we will walk you through what this means for you as a developer, and how using Neuton models can improve your development and your end application. 

Learn from the experts

On june 25th we're hosting a webinar where one of the lead developers in the Neuton team will lead you through the workflows of creating ultra-lightweight and ultra-fast Neuton models to enable Edge AI capabilities on any SoC.

Learn how to collect data to prepare a dataset and build a highly optimized Neuton model that can run on everything from the flagship nRF54L15 SoC to the most constrained devices in the Nordic lineup, like the nRF52805 SoC. After the hands-on presentation, you will be able to ask questions directly to the team behind developing the technology that made it possible in a live Q&A session.

Benchmarks

"Magic Wand" gesture recognition

 Total Footprint LiteRT Neuton Neuton Advantages 
NVM
 TinyML framework (model + inference engine + DSP) 
 79.96 5.42 14 times smaller model
43% reduction of total NVM use 
Device drivers and business logic 93.47 93.47
RAM TinyML framework (model + inference engine + DSP) 18.2 1.72 10 times smaller model
26% reduction of total RAM use 
Device drivers and business logic 45.69 45.69
Inference time (µs) 55,262 1,640 33 times faster 
Holdout validation accuracy 0.93 0.94  0.7% higher accuracy

Test performed with both Neuton and LiteRT models running on an nRF52840 and tested on the same holdout dataset.

    Use cases

  • Predictive maintenance and building automation systems

    In building automation systems, embedded machine learning can be leveraged to monitor equipment health in real-time and anticipate failures before they occur. By analyzing sensor data directly on edge devices, these systems can detect patterns and anomalies in HVAC, lighting, and security operations without relying on constant cloud connectivity. This approach reduces downtime, optimizes energy use, and lowers maintenance costs, all while enabling smarter, more responsive building environments.

  • Smart sensor networks with local data analysis on each node

    Smart sensor networks can utilize embedded machine learning to process information directly at the source, reducing latency and bandwidth requirements. Each node can independently detect patterns, filter noise, and make real-time decisions, enabling more efficient and scalable systems. This decentralized approach enhances responsiveness and reliability, especially in applications like environmental monitoring, industrial automation, and smart buildings, where immediate insights and minimal data transmission are critical.

  • Movement and gesture recognition for remote controls and wearable devices

    For remote controls and wearable devices, embedded machine learning enables real-time interpretation of motion data directly on the device. By running lightweight ML models on embedded processors, these devices can accurately detect and classify user gestures without needing constant cloud access. This allows for intuitive, low-latency interactions in applications like touchless control, fitness tracking, and assistive technologies, while maintaining energy efficiency and data privacy.

  • Health and activity monitoring for smart health wearables

    Smart health wearables are increasingly powered by embedded machine learning, enabling real-time analysis of biometric and motion data directly on the device. These wearables can track vital signs, detect irregularities, and classify physical activities with minimal latency. By processing data locally, they improve user privacy, extend battery life, and provide immediate feedback, making them ideal for continuous, personalized health monitoring and early detection of potential medical issues.

Demo videos