Haidian AI Enterprise Sprints Towards Being the "First General Decision-Making Large Model Stock"
SOURCE: Beijing Haidian
TIME: 2026.06.10
On June 9, the Hong Kong Exchanges and Clearing Limited (HKEX) website showed that Beijing Zhongke WengAI Science and Technology Co., Ltd. (hereinafter referred to as "Zhongke WengAI"), an artificial intelligence enterprise in Haidian District, has published its post-hearing information pack, meaning the company has officially passed its listing hearing. As the listing process advances, the Hong Kong stock market may welcome a hardcore AI company from Haidian centered on the "General Decision-Making Large Model".
The prospectus shows that Zhongke WengAI was founded in 2017 by a team of scientists from the Institute of Automation, Chinese Academy of Sciences, focusing on the research and development of enterprise-level large model-driven decision intelligence operating systems and services. According to CIC Consulting data, by 2025 revenue, Zhongke WengAI ranked first among enterprise-level large model-driven decision intelligence service providers in China, with a market share of 10.2%. 
Since its founding, the team has long been deeply engaged in complex information analysis, cognitive intelligence, social computing, multi-agent systems, and AI-assisted decision-making, and has continuously implemented solutions in complex business scenarios such as public governance, finance, media convergence, and industrial intelligence, gradually accumulating core capabilities in data analysis, business ontology modeling, intelligent assessment, and multi-agent deduction. Recently, the company released Decitron, a General Decision-Making Large Model built on full-stack self-developed AI technology, making "General Decision-Making Large Model" the new keyword for the outside world to understand its decision intelligence business. 
From the information disclosed in the prospectus, Zhongke WengAI’s core business has long been centered on "decision intelligence". The company provides decision support services to government and enterprise clients through its self-developed DIOS (Decision Intelligence Operating System). Financial data shows that its revenue grew from RMB 249.7 million in 2023 to RMB 405.3 million in 2025, an increase of over 60%. Gross margin improved from 44.0% in 2023 to 51.2% in 2025, and has been stably maintained at around 50%. 
Data shows that the company has cumulatively provided professional AI services to over 650 enterprise and government clients. In 2023, 2024 and 2025, the number of clients served in each year was 262, 342 and 404 respectively, with the number of commercial enterprise clients growing from 131 in 2023 to 184 in 2025. 
The prospectus shows that the company plans to further expand AI applications to commercial, industrial and other fields, and continuously incubate new scenarios in science and education, energy and sustainable development, and medical health. The company also plans to launch cloud-based DIOS (Decision Intelligence Operating System) services to further enhance product standardization and replicability. In terms of international market expansion, the company will initially focus on Hong Kong, the Middle East and Southeast Asian markets. 
Prior to the IPO, the company has completed 10 rounds of financing, with investors including national-level funds such as the China Development Bank National Manufacturing Transformation and Upgrading Fund, China Internet Investment Fund, and CCTV Media Convergence Industry Investment Fund, as well as investment institutions including Beijing AI Industry Investment Fund, Zhongguancun Science City Sci-Tech Growth Fund, CASSTAR, SCGC and Infotech.
It is reported that against the backdrop of AI applications evolving from content generation to decision intelligence, Zhongke WengAI's passage of the HKEX hearing has also brought the market label of "First General Decision-Making Large Model Stock" into the capital market's view, making it a rare asset that combines technological accumulation, product systems and industry implementation capabilities. 
返回上一页