Bingo Software (SH:688227) Quan Jing Wang·2025-10-27 10:00Core Insights - The 27th China International Software Expo concluded in Zhengzhou, focusing on AI reshaping software and industrial software innovation, with significant participation from leading companies and industry experts [1] - Pingao Co., Ltd. showcased its self-developed Pingyuan AI integrated machine, winning the "2025 Annual Artificial Intelligence Innovation Product" award, marking a breakthrough in domestic AI computing power facilities [1] Industry Overview - The demand for computing power is surging due to the rapid adoption of generative AI and large model technologies, with projections indicating that China's intelligent computing power will reach 1037.3 EFLOPS by 2025, a 43% increase [2] - The China Academy of Information and Communications Technology forecasts that by 2028, China's intelligent computing power will reach 2782 EFLOPS, with a compound annual growth rate exceeding 40% over the next five years [2] - Domestic AI computing power construction faces challenges, including high costs of traditional GPU clusters and reliance on imported core hardware, which poses risks to supply chain stability [2] Product Innovation - The Pingyuan AI integrated machine, launched in July 2025, offers a fully domestic solution with high-density computing and low total cost of ownership (TCO), addressing industry pain points [3] - The product features a fully localized supply chain design, supporting 16 Jiangyuan D10/D20 AI inference cards, and is compatible with major CPUs and domestic operating systems [3] - The machine achieves 5 PetaOPS computing power in a 4U standard rack space, significantly enhancing response speed for large models and meeting high-frequency application needs [4] Market Recognition and Future Strategy - Winning the software expo award signifies the technical strength and application value of the Pingyuan AI integrated machine, with a focus on local deployment and data security [5] - The company aims to drive the "vertical field AI + domestic computing power ecosystem" strategy, increasing R&D investment and collaborating with educational and research institutions to build a domestic computing power ecosystem [5]