21世纪经济报道·2025-09-05 00:14Core Viewpoint - The domestic AI chip industry is experiencing significant growth driven by strong demand for AI inference, with leading companies accumulating finished products and reserving key materials to prepare for future market developments [1][4]. Group 1: Company Performance - Haiguang Information achieved a revenue of 5.464 billion yuan in the first half of the year, a year-on-year increase of 45.21%, and a net profit of 1.201 billion yuan, up 40.78% [4]. - Cambricon Technologies reported a staggering revenue growth of 4347.82%, reaching 2.881 billion yuan, and a net profit of 1.038 billion yuan, compared to a loss of 530 million yuan in the previous year [4]. - Longxin Zhongke's revenue was 244 million yuan, a 10.9% increase, but it still faced a net loss of 294 million yuan, which is a 23.53% decline compared to the previous year [4]. Group 2: Inventory and Contracts - Cambricon's inventory increased by 51.64% to 2.69 billion yuan, accounting for 31.95% of total assets, while contract liabilities surged by 61223.22% to 543 million yuan, representing 6.45% of total assets [5][6]. - Haiguang Information's contract liabilities rose by 242.1% to 3.091 billion yuan, making up 9.57% of total assets, with inventory reaching 6.013 billion yuan, a 10.84% increase [6]. Group 3: Market Position and Competition - Domestic AI chip manufacturers are still in a position to increase their market share significantly, with various companies adopting different technological routes to compete [1][9]. - In the Chinese AI acceleration chip market for 2024, Nvidia holds a 66% market share, while domestic players like Huawei HiSilicon and Muxi account for approximately 23% and 1%, respectively [9]. - ASIC customized chips are gaining traction due to their tailored design for specific AI inference scenarios, attracting interest from major cloud computing companies [9]. Group 4: Ecosystem Development - The development of a collaborative and open industry chain environment is crucial for the sustained growth of the domestic AI chip ecosystem [13][16]. - Companies like Haiguang Information are focusing on deep collaboration across the industry chain to enhance the integration of hardware, software, and ecosystem [15]. - The rapid advancement of domestic AI chips is attributed to years of sustained R&D investment and a collective effort within the industry to foster an open ecosystem [16].