nRF54LM20B – New ultra-low-power, large-memory wireless SoC with Axon NPU;
Custom Neuton Models – Industry-leading, ultra-compact edge AI models;
Nordic Edge AI Lab – Development tool that simplifies and accelerates edge AI development.
“AI factories train intelligence, but Nordic deploys it: on the device, at the edge, where the world happens,” said Vegard Wollan, CEO of Nordic Semiconductor. “Edge AI is no longer optional: it’s the only way to deliver security, privacy, and sustainability at scale. Nordic’s edge AI solution enables millisecond decisions with zero latency round-trip to the cloud, ensures regulatory compliance through local processing, and delivers radically improved battery life for billions of connected devices. This is the new standard for ultra-low-power edge AI, and Nordic is defining it.”

In 2023, Nordic acquired Atlazo and its Axon technology. The nRF54LM20B SoC is the first large-memory member of the nRF54L series and integrates the Axon Neural Processing Unit (NPU), an ultra-efficient AI hardware accelerator to power demanding edge AI workloads. Axon delivers up to 7x the performance and up to 8x the power efficiency of competing solutions in tasks such as sound classification, keyword detection, and image-based detection.
The nRF54LM20B combines the Axon NPU with 2MB of non-volatile memory, 512KB of RAM, a 128MHz Arm® Cortex-M33 plus a RISC-V coprocessor, high-speed USB, up to 66 GPIOs, and Nordic's ultra-low-power fourth-generation 2.4GHz radio supporting Bluetooth® LE, Bluetooth Channel Sounding, Matter over Thread, and more.

Neuton models are ultra-compact edge AI models that run on CPUs, typically under 5 KB and up to 10 times smaller, faster, and more efficient than other equivalent models. Nordic Edge AI Lab helps developers generate custom Neuton models for anomaly detection, activity and gesture recognition, biometric monitoring, and more, providing privacy-preserving real-time intelligence on devices with tiny batteries and limited memory, without relying on the cloud.

With Nordic Edge AI Lab and Neuton models, intelligence moves seamlessly from concept to reality. A recent deployment demonstrates this: a global supply chain solution upgraded its smart tracking devices with AI models created at Nordic Edge AI Lab, enabling the detection of real-world handling events such as impacts, shaking, and transport directly on an nRF54L Series SoC. This AI-driven insight was deployed across the entire fleet without operational disruption, thanks to Nordic's nRF Cloud lifecycle services.

“With Nordic Edge AI Lab, the Neuton models, and the Axon NPU, Nordic is making advanced on-device AI practical for any embedded developer,” said Oyvind Strom, EVP of Nordic Semiconductor’s short-range business unit. “Developers gain the simplicity to move quickly and the disruptive performance to scale from wearables to industrial sensors, all enabled by Nordic’s reliable, ultra-low-power hardware solutions.”
As intelligence moves to the edge, the need for over-the-air (OTA) management and deep observability grows. At the same time, cloud services remain critical for device management, embedded observability, and location services. Manufacturers increasingly need continuous visibility into device performance to improve products in real time and meet regulatory and customer requirements, ensuring that connected products evolve safely and efficiently throughout their lifecycle.

available
today for Nordic’s nRF54 Series Wireless SoCs and Cellular IoT SiPs. The nRF54LM20B SoC with Axon NPU is currently being sampled by select customers, with general availability for development expected in early Q2 2026.