Amid the high-performance GPU turf tussle between AMD and Nvidia (and soon, Intel), a new, China-based player is emerging: Biren Technology, founded in 2019 and headquartered in Shanghai. "It is my honor to present our first-generation compute product: BR100," Xu said. "BR100 is dedicated to addressing the challenges of AI training and inference in the datacenter, with augmented goals of increasing productivity and reducing overall cost of ownership." At 1074mm2, the 77 billion-transistor, dual-die Biren BR100 (pictured in the header) will be manufactured on TSMC's 7nm process and capable of 256 FP32 teraflops. The die-to-die interconnect provides 896 gigabytes per second bandwidth.