Nordic Semiconductor Unveils nRF71 Series, Slashing Standby Power and Bringing Edge AI to Battery-Powered Wi-Fi Devices

Nordic Semiconductor has introduced its new nRF71 series of wireless systems-on-chip (SoCs), marking a significant leap forward in low-power wireless connectivity. According to the company, the nRF71 series delivers up to 20 times lower standby power consumption compared to its predecessor line, the nRF70 series of Wi-Fi companion ICs. This substantial reduction in power draw represents a major milestone for developers seeking to incorporate Wi-Fi into compact, battery-powered electronic devices without sacrificing operational lifespan.

Both the nRF70 and the new nRF71 devices draw an average connected current of 14 microamps at 2.4 GHz and 8 microamps at 5 GHz when operating at DTIM 10, demonstrating high efficiency during active network connection states. However, the dramatic improvements in standby efficiency found in the nRF71 architecture open up entirely new operational parameters for hardware engineers designing portable and remote connected systems.

The nRF71 lineup currently features two primary variants designed to cater to distinct connectivity and processing requirements. The standard nRF7120 offers dual-band Wi-Fi 6, providing robust and efficient short-range wireless networking. Meanwhile, the nRF7120E expands upon this foundation by adding tri-band Wi-Fi 6E capabilities alongside an integrated Axon Neural Processing Unit (NPU) engineered specifically for hardware-accelerated edge artificial intelligence workloads.

Beyond their advanced Wi-Fi capabilities, both devices are designed to serve as comprehensive multi-protocol hubs. Each chip supports Bluetooth Low Energy, Thread, and Matter protocols, ensuring broad interoperability across various smart home and industrial ecosystems. To accommodate complex firmware, application code, and local data processing, the SoCs feature 4 megabytes of on-chip non-volatile memory alongside 1 megabyte of user-accessible RAM.

The dramatic reduction in ultra-low power consumption delivered by the nRF71 series is widely anticipated to enable a new generation of battery-powered Wi-Fi designs. Historically, integrating Wi-Fi into battery-operated equipment posed severe thermal and energetic challenges, frequently forcing developers to rely on lower-bandwidth alternatives like Bluetooth or Zigbee due to power constraints. With the nRF71 series, applications within the smart home sector, wearable technology market, and industrial Internet of Things (IoT) landscape can now leverage the high throughput and dense network capabilities of Wi-Fi 6 and Wi-Fi 6E while maintaining extended battery longevity.

Furthermore, the integration of the Axon NPU within the nRF7120E variant addresses one of the most pressing bottlenecks in modern connected device engineering: data privacy, latency, and power costs associated with cloud-based artificial intelligence. By incorporating hardware acceleration directly onto the chip, AI inference can run locally at the extreme edge. This eliminates the need for an external AI co-processor, thereby saving physical circuit board space, reducing overall system BOM costs, minimizing communication latency, and conserving critical battery reserves that would otherwise be wasted transmitting raw sensor data to the cloud.

Nordic Semiconductor highlights several practical edge AI use cases that are well-suited for battery-powered devices built around the nRF71 series. Among these are keyword spotting for voice-activated interfaces, anomaly detection in industrial machinery monitoring, and local image classification for smart security or environmental sensing equipment.

The performance gains enabled by the dedicated Axon NPU are substantial. Machine-learning models executed via the NPU run between 10 times and 17 times faster than they would running purely on a conventional 256-MHz Cortex-M33 processor core alone. This acceleration ensures that complex neural network tasks complete rapidly, allowing the device to return to its ultra-low-power sleep state almost immediately and preserving battery reserves.

In addition to hardware-accelerated processing via the Axon NPU, the nRF71 series offers flexible software-based alternatives for machine learning deployment. The devices are capable of executing ultra-small AI models that have been specifically trained on developers’ own proprietary data using Nordic’s Neuton software development tools. This capability enables edge AI inference to run directly on the standard processor architecture even without dedicated NPU hardware acceleration, giving development teams diverse options depending on their specific product tier and processing demands.

The introduction of the nRF71 series comes at a time when the market demand for smarter, more autonomous edge devices continues to accelerate across both consumer and industrial verticals. As smart home ecosystems mature through universal standards like Matter, the requirement for seamless, low-power, multi-protocol devices that can process local context intelligently has never been higher. By combining multi-band Wi-Fi 6 and 6E connectivity, robust multiprotocol support, generous memory allocations, and integrated edge AI processing into a single silicon platform, Nordic Semiconductor is positioning the nRF71 family as a versatile foundational architecture for future embedded designs.

Nordic Semiconductor has announced that the nRF71 series is currently sampling with selected customers as the company prepares for broader commercial availability. Detailed product documentation, technical specifications, and development resources for the nRF7120 and nRF7120E can be accessed directly through the Nordic Semiconductor website.