M31 Technology Corporation (M31), a global leader in silicon intellectual property (IP), made a prominent appearance at Cheng-Yu IC 2025 Annual Development Forum (ICCAD-Expo 2025) in Chengdu, China. During the event, M31 unveiled a series of cutting-edge IP technologies designed for artificial intelligence (AI) and low-power applications. From high-performance MIPI C/D-PHY RX interface solutions on N4 to N3, and low-power design IP on N12e, to the latest N6e ULL, eLL, and Low-VDD memory compilers, M31 showcased comprehensive technological capabilities in advancing innovation across smart computing, automotive electronics, and mobile devices—further reinforcing the company’s commitment to driving the next generation of AI chip design.

This year’s exhibition, themed “Cheng-Yu One Chip, Resonating at the Same Frequency,” highlighting innovation and collaboration within the IC design industry. As AI-driven robotics and autonomous systems continue to evolve, M31 introduced high-performance MIPI C-PHY v2.1 (6.5 Gsps) and D-PHY v3.0 (9 Gbps) PHY solutions on the TSMC N3 process platform. These solutions enable robotics and autonomous driving applications to fully perceive their surroundings and make real-time decisions, meeting the high-bandwidth, low-latency demands of imaging and sensing systems to achieve high-speed data transmission and precise environmental recognition.

For automotive ADAS and high-resolution imaging applications, M31 also launched the N4 MIPI C-PHY v2.0 (6.0 Gsps) and D-PHY v2.1 (4.5 Gbps) solutions, supporting more efficient image streaming and data processing. These technologies accelerate the integration of intelligent driving and automotive electronic systems while providing robust and flexible high-speed interface options for automotive chip design and have already been adopted by leading EV manufacturers.

At the same time, to drive wider AI adoption in edge computing devices, M31 has launched ultra-low-power (ULL), extremely-low-leakage (eLL), and low-voltage (Low-VDD) memory compilers on the TSMC N6e and N12e platforms, supporting Always-On blocks and wide-range operating voltage. Developed using TSMC’s advanced N6e process, this series of memory compilers is purpose-built for AI edge computing and Internet of Things (IoT) devices, delivering high density, high performance, and exceptionally low power consumption—enabling systems to retain AI inference capabilities while extending battery life. In particular, the ULL compiler supports Dynamic Voltage and Frequency Scaling (DVFS) and High Sigma design, ensuring stable operation under stringent conditions. The eLL variant reduces power consumption by up to 50% in deep sleep mode, while the Low-VDD design operates at voltages as low as 0.5 V, balancing both performance and energy efficiency. This comprehensive product portfolio delivers a rare low-power IP solution for edge AI applications, showcasing M31’s innovation and technological strength in next-generation intelligent computing and SoC integration.

At this year’s ICCAD exhibition, M31 focused on AI, automotive electronics, and low-power applications, unveiling a series of high-performance IP achievements that further strengthen its leadership in AI and automotive chip innovation. M31 was also honored for the eighth consecutive time with TSMC’s OIP “Specialty Process IP Award,” underscoring the strong recognition from the world’s leading foundry for M31’s technical excellence and long-term partnership.

M31 CEO Scott Chang attended the event in person and stated, “Our long-standing collaboration with leading advanced foundries has been a cornerstone of M31’s continued innovation. We remain committed to delivering IP solutions that balance performance and power efficiency, helping AIoT, automotive electronics, and edge computing applications accelerate successful chip design and implementation. Looking ahead, we look forward to deepening our cooperation with partners in China’s IC design industry, jointly advancing AI-driven technologies and driving further industrial upgrading.”