[フレーム]

Upbeat and SiFive to Demonstrate Dual-Core RISC-V Microcontrollers with AI Acceleration at RISC-V Summit

By Chad Cox

Production Editor

Embedded Computing Design

October 20, 2025

News

Image Credit: Upbeat Technology

Upbeat Technology and SiFive, Inc. introduced the UP201/UP301 family MCU, a dual-core RISC-V microcontroller engineered for ultra-low power efficacy. The companies plan to exhibit the solutions at the upcoming RISC-V Summit in Santa Clara, California from October 22–23, 2025.

The UP201/UP301 MCUs integrates two SiFive Essential IP Cores and two AI accelerators designed by Upbeat, enabling market-leading performance per watt while supporting always-on IoT devices, wearables, drones, and edge AI inference.

Due to its combination of near‐threshold operation, Error Detection and Correction (EDAC) resilience, and SRAM innovation, the MCUs deliver energy efficiency at 16.8 uW/MHz/DMIPS (Dhrystone).

When juxtaposed with general ARM‐based microcontroller architectures, the MCU portfolio enhances design with a scalable (up to 400 MHz or 717 DMIPS) production‐ready platform that balances performance, efficiency, and intelligence for next‐generation connected devices.

“From bring-up to production, Upbeat Technology’s UP201/UP301 family of MCUs has been very easy to work with. Its power profiles exceeded our targets—ultra-low power in sleep and highly efficient under active load. The integrated AI accelerator and DSP offload inference and signal processing, enabling us to meet real-time deadlines. Thanks to the SoC’s low-power, high-performance architecture, we can push more on-device processing—richer models, longer duty cycles, and new features—without compromising battery life,” said FuJen Hsiao, CEO, Awan-sensing.

Visitors will also be able to see Upbeat Technology’s ‘A Heterogeneous Dual-Core RISC-V Edge AI Platform for Power-Efficient Machine Learning’ on October 22, 2025, at 11:10 a.m., featuring live demos and hands-on training with the UP201/UP301 family MCU.

For more information, visit sifive.com and upbeattechtw.com.

Subscribe

Chad Cox. Production Editor, Embedded Computing Design, has responsibilities that include handling the news cycle, newsletters, social media, and advertising. Chad graduated from the University of Cincinnati with a B.A. in Cultural and Analytical Literature.

More from Chad

Categories
Open Source - RISC-V & Open Source IP
Processing - Chips & SoCs
Processing - Compute Modules
AI & Machine Learning
AI & Machine Learning - AI Development Tools & Frameworks
AI & Machine Learning - AI Logic Devices & Workload Acceleration
Topic Tags
Automotive
MORE
Networking & 5G
MORE
Software & OS
MORE
HPC/Datacenters
MORE

AltStyle によって変換されたページ (->オリジナル) /