Wave Computing, founded in 2010 and based in Campbell, California, specializes in developing hardware solutions for artificial intelligence and machine learning applications. The company's core technology is centered around a patented dataflow architecture that eliminates the need for traditional central processing units (CPUs) or graphics processing units (GPUs). This approach aims to remove performance and scalability bottlenecks typically associated with conventional deep learning solutions.
Wave Computing's flagship product is a compute appliance designed to accelerate machine learning applications. As of 2017, the company claimed that its Deep Learning Computers could deliver up to 1000 times the performance for neural network training compared to traditional solutions. The company offered an early access program for interested parties, with general availability of these computers planned for 2017.
In June 2018, Wave Computing acquired MIPS, a company known for its processor architecture. Following this acquisition, Wave Computing outlined its strategy for an AI-enabled MIPS offering and ecosystem. The company announced plans to expand its IP team globally, including hardware, software, marketing, and sales staff. Wave Computing committed to continuing investment in the 64-bit, scalable multi-threaded MIPS technology roadmap for embedded applications.
Wave Computing's product strategy included offering solutions ranging from CPU cores for edge applications to more robust implementations for emerging AI training and inferencing applications. The company also addressed the future of functional safety in autonomous vehicles by building on its ISO26262 certification and introducing advanced lock-step functionality for its MIPS cores.
By using this site, you agree to allow SPEEDA Edge and our partners to use cookies for analytics and personalization. Visit our privacy policy for more information about our data collection practices.