Specifications
Product Class
AI-native connected microcontroller (MCU)
MCU Core
Arm Cortex-M52 at 200 MHz with Arm Helium vector DSP extensions
NPU
Arm Ethos-U55, up to 50 GOPS at 200 MHz
AI Capability
On-device inference; Wi-Fi sensing, sound event detection, wake-word, contextual awareness
Wireless
Tri-band Wi-Fi 7 (2.4 / 5 / 6 GHz), 1×1 20 MHz
Bluetooth
Bluetooth LE 6.0 with channel sounding for accurate distance measurement
802.15.4
Thread and Zigbee, with Matter interoperability
Concurrent Radios
Simultaneous Wi-Fi, Bluetooth and Thread operation
RF Integration
Integrated LNA, PA and T/R switches
Throughput / Range
Sustained 20 Mbps to 200 m
Footprint
PCB area under 100 mm², QFN package enabling non-HDI PCB
BOM Impact
Up to 25% lower RBOM than comparable multi-chip solutions
Security
On-chip security architecture for connected device provisioning
Overview
The Synaptics SYN765x is the industry's first Wi-Fi 7 AI-native MCU: a single chip that combines an Arm Cortex-M52 application core running at 200 MHz with Helium vector DSP extensions, an Arm Ethos-U55 NPU delivering up to 50 GOPS, and a tri-band Wi-Fi 7 radio with Bluetooth 6.0 and IEEE 802.15.4 (Thread / Zigbee) on 2.4, 5 and 6 GHz.
The architectural point is that the accelerator is on the same die as the radio and the MCU. Connected devices have historically streamed raw data to a host or the cloud for inference; the SYN765x runs Wi-Fi Channel State Information extraction and machine-learning models locally, so presence detection, motion tracking, proximity sensing and sound event detection happen on the device and only significant events travel over the network. That removes both the bandwidth cost and much of the latency of cloud round-trips.
Integration is the second differentiator. Bluetooth Channel Sounding gives accurate distance measurement using standard radios rather than mmWave radar or ultra-wideband, and the integrated LNA, PA and transmit/receive switches bring the total PCB footprint under 100 mm² in a QFN package that avoids HDI board technology. Synaptics cites up to 25% lower rolled-up BOM cost against comparable multi-chip designs.
Key Benefits
Industry-first Wi-Fi 7 AI-native MCU — MCU, NPU and radio on one die. 50 GOPS Ethos-U55 NPU for on-device inference without cloud round-trips. Tri-band Wi-Fi 7 + Bluetooth 6.0 + Thread/Zigbee/Matter with concurrent operation. Under 100 mm² on a non-HDI QFN PCB, with up to 25% lower RBOM.
Applications
Smart-home and building sensors with on-device presence and motion detection, security cameras performing local event classification, battery-powered voice and audio devices, industrial wireless sensing nodes, and Matter-based IoT products that need local inference at low power.
Request a Quote — SYNAPTICS SYN765X — WI-FI 7 AI-NATIVE CONNECTED MCU WITH ETHOS-U55 NPU
QS Compute — global B2B supply of AI computing hardware, edge AI systems and accelerators. Volume pricing, 15-day sample lead time.
Get Your Quote →