Preferred Networks adopts Siemens’ software for next-gen AI chip design

Founded in 2014, PFN is known for research and development in areas like machine learning, robotics, and autonomous systems. The company collaborates with innovators across a range of industries, including manufacturing, healthcare, and transportation, to integrate AI solutions into their operations.

PFN used the software’s broad array of power-saving register transfer-level (RTL) features to develop a new generation of AI integrated circuits (ICs) that deliver exceptional performance and low power consumption. Siemens’ PowerPro software also enabled PFN to rapidly conduct accurate power estimation for both RTL and gate-level designs, allowing its design teams to quickly find and address power issues during RTL development – helping to dramatically enhance the power efficiency of its next generation AI semiconductor designs.

“Siemens’ PowerPro software enables us to design low-power and energy efficient AI chips that meet the stringent standards that our customers demand,” said Junichiro Makino, Chief Technology Officer for Computer Architecture at PFN, and Professor-at-Large at Kobe University. “This advanced technology allows us to take our project to the next level, while helping our customers develop compelling and highly differentiated end-products based on our next-generation AI semiconductors.”

Based on deep sequential analysis technology, it rapidly identifies the most effective clock and memory gating pathways for the design, which can dramatically boost overall power efficiency.

“Siemens is pleased that PFN has adopted our PowerPro software to enhance the power efficiency of their next-generation AI processor designs,” said Ankur Gupta, Senior Vice President and General Manager, Digital Design Creation Platform, Siemens Digital Industries Software. “PowerPro improves the power efficiency of next-generation designs and we look forward to the innovation that PFN’s next generation AI processors will unleash.”