NEUCHIPS secures funding to deliver deep learning AI inference
NEUCHIPS, a leader in AI ASIC platforms for deep learning recommendation, has announced series B2 funding of $20M. The round includes strategic investors Wistron, JAFCO Asia, Powerchip Group and new investor Raydium Semiconductor Corporation.
Leading cloud service providers have shared that the overwhelming majority (over 80%) of the AI inference cycles in the data center are actually for recommendation models, not vision or language. The first product from NEUCHIPS, RecAccel™ N3000, was designed specifically for Deep Learning Recommendation Models (DLRM) and will be ready to sample later this year with easy-to-use software and leading-edge performance.
RecAccel™-N3000 has achieved one million DLRM inferences per Joule of energy (which translates into 20 million inferences per second per 20-Watt chip). The AI accelerator, developed with support and EDA tools from Synopsys and other semiconductor industry leaders, will be manufactured on a 7nm TSMC process.
“NEUCHIPS’ focus on business integrity, technology innovation, talent development and value co-creation aligns with our vision,” said Hermit Huang, Chairman of Raydium Semiconductor Corporation. “We are confident this investment will continue to create synergies between Raydium and NEUCHIPS.”
“The NEUCHIPS deep learning recommendation ASIC is setting the standard for high performance and low power. We are confident that NEUCHIPS is on the right track to deliver the most energy-efficient and cost-effective AI inference solution to impact the market,” said John Lin, Managing Director, Head of Taiwan JAFCO Asia.
“I am optimistic about NEUCHIPS’s promising future, and look forward to collaborating with them on new technology such as our new specialty memory and Wafer-on-Wafer(WoW) technology to boost performance per watt and performance per dollar,” said Frank Huang, the Founder of Powerchip Group.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
