Renesas Pushes Forward on Edge AI and Power Management with Two Announcements

July 01, 2025

News

To bring focus onto a pair of major announcmeents for Renesas, the company hosted a virtual press conference to announce its new single- and dual-core MCUs for Edge AI, and new GaN FETs for high-density power conversion in AI data centers, industrial and charging systems.

The new Renesas RA8P1 microcontroller (MCU) Group is reportedly targeted at AI and ML applications. The new MCUs combine 1GHz Arm Cortex-M85 and 250MHz Cortex-M33 CPU cores with the Arm Ethos-U55 Neural Processing Unit (NPU), said Daryl Khoo during the virtual press conference. This combination is designed to deliver high CPU performance at over 7300 CoreMarks and AI performance of 256 GOPS at 500 MHz, he said.

The Ethos-U55 NPU is used to offload the CPU for compute intensive operations in Convolutional and Recurrent Neural Networks (CNNs and RNNs) to deliver up to 256 MACs per cycle and it supports most commonly used networks, including DS-CNN, ResNet, Mobilenet TinyYolo and others, the company said. Depending on the neural network used, the Ethos-U55 reportedly provides up to 35x more inferences per second than the Cortex-M85 processor on its own.

The RA8P1 MCUs are manufactured on the 22ULL (22nm ultra-low leakage) process from TSMC, which enables the use of embedded Magnetoresistive RAM (MRAM) in the new MCUs to offer faster write speeds and higher endurance and retention than is available with Flash memory.

Along with the RA8P1 MCUs, Renesas has introduced RUHMI (Renesas Unified Heterogenous Model Integration), a comprehensive framework for MCUs and MPUs. RUHMI has been built to offer efficient AI deployment of the latest neural network models in a framework agnostic manner, Renesas said. It enables model optimization, quantization, graph compilation and conversion, and generates efficient source code. RUHMI provides native support for machine-learning AI frameworks such as TensorFlow Lite, Pytorch & ONNX. It also provides the necessary tools, APIs, code-generator, and runtime needed to deploy a pre-trained neural network, including ready-to-use application examples and models optimized for RA8P1. RUHMI is integrated with Renesas’s own e2Studio IDE to allow seamless AI development. This integration will facilitate a common development platform for MCUs and MPUs.

Key Features of the RA8P1 MCUs

• Processors: 1GHz Arm Cortex-M85, 500MHz Ethos-U55, 250 MHz Arm Cortex-M33 (Optional)
• Memory: 1MB/512KB On-chip MRAM, 4MB/8MB External Flash SIP Options, 2MB SRAM fully ECC protected, 32KB I/D caches per core
• Graphics Peripherals: Graphics LCD controller supporting resolutions up to WXGA (1280x800), parallel RGB and MIPI-DSI display interfaces, powerful 2D Drawing engine, parallel 16bit CEU and MIPI CSI-2 camera interfaces, 32bit external memory bus (SDRAM and CSC) interface
• Other Peripherals: Gigabit Ethernet and TSN Switch, XSPI (Octal SPI) with XIP and DOTF, SPI, I2C/I3C, SDHI, USBFS/HS, CAN-FD, PDM and SSI audio interfaces, 16bit ADC with S/H circuits, DAC, comparators, temperature sensor, timers
• Security: Advanced RSIP-E50D cryptographic engine, TrustZone, Immutable storage, secure boot, tamper resistance, DPA/SPA attack protection, secure debug, secure factory programming, Device Lifecycle management
• Packages: 224BGA, 289BGA

The RA8P1 MCUs are available now. Renesas is also shipping an RA8P1 Evaluation Kit. Samples and kits can be ordered either on the Renesas website or through distributors.

GaN Power Systems

Renesas at the same time introduced three new high-voltage 650V GaN FETs for AI data centers and server power supply systems including the new 800V HVDC architecture, E-mobility charging, UPS battery backup devices, battery energy storage and solar inverters. Designed for multi-kilowatt-class applications, these 4th-generation plus (Gen IV Plus) devices combine high-efficiency GaN technology with a silicon-compatible gate drive input, significantly reducing switching power loss while retaining the operating simplicity of silicon FETs, according to Primit Parikh of Renesas during the press conference. Offered in TOLT, TO-247 and TOLL package options, the devices are designed to give engineers the flexibility to customize their thermal management and board design for specific power architectures.

The new TP65H030G4PRS, TP65H030G4PWS and TP65H030G4PQS devices leverage the SuperGaN platform, a depletion mode (d-mode) normally-off architecture by Transphorm, which was acquired by Renesas in June 2024. Based on low-loss d-mode technology, the devices reportedly minimize power loss with lower gate charge, output capacitance, crossover loss, and dynamic resistance impact, with a 4V threshold voltage.

Built on a die that is 14 percent smaller than the previous Gen IV platform, the new Gen IV Plus products achieve a lower RDS(on) of 30 milliohms (mΩ), reducing on-resistance by 14 percent and delivering a 20 percent improvement in on-resistance output-capacitance-product figure of merit (FOM), the company said. The smaller die size reduces system costs and lowers output capacitance, which results in higher efficiency and power density. 

Available in compact TOLT, TO-247 and TOLL packages, they provide one of the broadest packaging options to accommodate thermal performance and layout optimization for power systems ranging from 1kW to 10kW, and even higher with paralleling. The new surface-mount packages include bottom side (TOLL) and top-side (TOLT) thermal conduction paths for cooler case temperatures, allowing easier device paralleling when higher conduction currents are needed. Further, the commonly used TO-247 package provides customers with higher thermal capability to achieve higher power.

The TP65H030G4PRS, TP65H030G4PWS and TP65H030G4PQS are available today, along with the 4.2kW Totem-pole PFC GaN Evaluation Platform (RTDTTP4200W066A-KIT).