Overview
Description
The RA8P1 group features Renesas’s first 32-bit AI-accelerated microcontrollers (MCUs) powered by the high-performance Arm® Cortex®-M85 (CM85) with Helium™ MVE and Ethos™-U55 NPU. It delivers 256GOPS* of AI performance, a breakthrough CPU performance of over 7300 CoreMarks and advanced artificial intelligence (AI) capabilities enabling voice, vision, and real-time analytics AI applications on a single chip. The RA8P1 MCUs built on the advanced 22nm ULL process are available in single and dual-core options with a Cortex-M33 core embedded on the dual-core MCUs.
The RA8P1 MCUs integrate high-performance CPU cores with large memory, multiple external memory interfaces, and a rich peripheral set optimized for AI applications. Single and dual-core devices in 224 and 289 BGA packages are available, to address diverse use cases across broad markets. Secure Element-like functionality is built-in with advanced cryptographic security IP, immutable storage, and tamper protection, to deliver truly secure Edge AI and IoT applications.
To streamline and accelerate development, the RA8P1 MCUs are supported by the Flexible Software Package, a comprehensive set of hardware and software development tools, and RUHMI - a highly optimized AI software platform providing all necessary tools for AI development, model optimization, and conversion, fully integrated with e2 studio.
*Giga Operations Per Second
Features
- 1GHz Arm Cortex-M85 and 250MHz Cortex-M33 cores
- Ethos-U55 NPU, 256GOPS at 500MHz
- 0.5/1MB MRAM, 4/8MB Flash options
- 2MB SRAM including TCM and 64KB caches
- Parallel and MIPI-CSI2 camera interface (I/F) for vision AI
- I2S and PDM I/F for voice AI
- GLCDC w/ parallel RGB and MIPI-DSI I/Fs
- 2D drawing engine
- xSPI compliant octal SPI with XIP & DOTF
- Renesas Security IP, TrustZone, tamper protection
- Secure boot with immutable storage for first stage bootloader
- 32/16-bit high-resolution timers, 32-bit ultra-low-power timer
- 16-bit ADC, 12-bit DAC, HS comparators
- Gigabit Ethernet, TSN switch, USB 2.0 HS/FS, CAN FD
- SDHI, SPI, I3C, I2C serial interfaces
- 32/16-bit external memory I/F (CS/SDRAM)
Comparison
Applications
Design & Development
Software & Tools
Sample Code
Boards & Kits
Models
ECAD Models
Schematic symbols, PCB footprints, and 3D CAD models from SamacSys can be found by clicking on the CAD Model links in the Product Options table. If a symbol or model isn't available, it can be requested directly from SamacSys.
Support
Support Communities
Renesas Academy
Support Communities
-
EK-RA8P1 SIP IS25WX064-JWLE
HI From the RA8P1 datasheet i found that, RA8P1 contains an internal OSPIO flash memory. In one of my project, I would like to use this ospi internal flash and external flash, but i couldn't able to find sample code for interfacing on chip SIP flash(IS25WX064-JWLE). My ...
Aug 5, 2025 -
Running AI Sample Application on EK-RA8P1 Board
Hello,I am trying to run one of the sample applications listed at the following GitHub repository on my EK-RA8P1 board: github.com/.../application_examples The project builds successfully, but when I run it on the actual hardware, the LCD screen remains dark and does not update ...
Jul 31, 2025 -
Unable to debug / program RA8P1 in e2 studio on Linux
Hi, I just recently got my hands on the new EK-RA8p1 development board and have been unsuccessful in debugging / programming it on the latest version of e2 studio and fsp version 6.0. I'm able to program other RA series microcontrollers (RA8M1, RA6E2) using the same ide ...
Jul 15, 2025
FAQs
-
How do I convert models for use with Ethos-U55 on RA8P1, using RUHMI?
When using Ethos-U55 on RA8P1, the model optimization and conversion process is carried out in several steps using the RUHMI Framework tool. Model Import: First, the user imports models from TensorFlow or ONNX format into the RUHM tool. RUHMI then converts these models into Intermediate Representation (IR) for further ...
Jul 1, 2025 -
How can I profile NPU inference performance and memory usage on RA8P1?
Arm Ethos-U55 includes a dedicated Performance Monitoring Unit (PMU) that enables evaluation of internal NPU operations. This allows developers to monitor various performance metrics such as bus access patterns and inference execution time. To perform performance profiling, developers can implement PMU control software, using the Ethos-U Core Software ...
Jul 1, 2025 -
RA8P1: How do I optimize memory resources and bandwidth for optimal performance with NPU?
The RA8P1 MCUs have large on-chip memory (1MB MRAM and 2MB SRAM) as well as external memory interfaces, that when used efficiently, can help to optimize the performance of the neural network model. Users typically use internal and external memories for storage of the AI model, input images, graphics ...
Jul 1, 2025
Videos & Training
Discover the AI-Accelerated RA8P1 microcontroller, powered by the Arm® Cortex®-M85 and Ethos™-U55 NPU, designed for energy-efficient AI applications in vision, voice, and real-time analytics. This high-performance microcontroller enables users to build advanced AI solutions with ease.
Learn about its key features and how the RUHMI Framework simplifies AI application development by allowing seamless deployment of neural network models in a framework-agnostic way. Watch now to explore how the RA8P1 MCU can transform your AI projects.
Related Resources
Additional Videos
Events & Webinars
News & Blog Posts
Blog Post
Aug 22, 2025
|
News
Jul 1, 2025
|
Blog Post
Jul 1, 2025
|
Blog Post
Dec 12, 2023
|
Blog Post
Oct 31, 2023
|
Blog Post
Oct 13, 2022
|
Blog Post
Jan 28, 2022
|