On April 12, Hong Kong Exchanges and Clearing disclosed that Beijing-registered Zhongke WengAI, also known as Wenge Technology, had confidentially filed a prospectus for a Hong Kong listing, with China International Capital Corporation (CICC) serving as its sole sponsor.
If approved, WengAI would become the first publicly listed artificial intelligence company to emerge from the Chinese Academy of Sciences ecosystem, according to the company. It would also be the first public company centered on decision-making large models. On April 10, the China Securities Regulatory Commission disclosed a filing notice related to WengAI’s overseas offering and listing, as well as the full circulation of its domestic unlisted shares.
Founded in 2017, WengAI was established by Wang Lei, Luo Yin, and Zeng Dajun, three scientists from the Institute of Automation at the Chinese Academy of Sciences. It has also assembled a team of young scientists from institutions including the Chinese Academy of Sciences, Tsinghua University, and Peking University.
Focused on complex data analytics and AI-assisted enterprise decision-making, WengAI said it has developed its own core AI capabilities and a full-stack suite of AI services.
At the center of that system is WengAI’s self-developed decision intelligence operating system, or DIOS, which underpins both its R&D framework and its AI services. The company said the system integrates a range of frontier technologies and is designed to turn raw data into business insights with practical value.
Built on data infrastructure, a reasoning platform that combines large models with multiple agents, and a low-code systems integration module, DIOS supports AI services with both broad applicability and industry-specific depth. According to the company, it enables customers to build AI applications more efficiently, process large multimodal datasets in complex business settings, and improve both operating efficiency and decision-making.
WengAI also said its self-developed Yayi large model ranks among the top performers in several mainstream industry benchmarks, with strong performance in Chinese semantic processing, AI agent task orchestration, and cross-scenario generalization.
The company also introduced the Yayi AI scientist agent, which has reportedly demonstrated strong complex reasoning capabilities. It said the agent ranked first on several core indicators in international evaluations, including GAIA, SimpleQA, and HLE, which measure scientific tool use and problem-solving ability.
Beyond the model layer, WengAI has expanded into both the data and decision layers. At the data layer, its enterprise-grade data integration and analytics platform, X-Data, is designed to address data silos through multimodal storage, dynamic orchestration, and multi-cloud compatibility. At the decision layer, its AI agent development platform, DI-Brain, aims to reduce coding requirements, lower technical barriers, and accelerate AI deployment across industries.
The company said the full-stack architecture of DIOS reduces complexity and integration challenges in multi-system environments, lowering both upfront deployment and ongoing maintenance costs. Between 2023–2025, WengAI’s average delivery cycle fell from 185 days to 80.2 days.
Between 2023–2025, WengAI said it provided AI services to more than 650 enterprise and government clients, particularly in China’s tech-intensive industries. Its stated goal is to help organizations in media and communications, public services, and commercial enterprise achieve AI-enabled digital transformation.
According to China Insights Consultancy, WengAI ranked first among China’s enterprise decision intelligence service providers powered by large models by 2024 revenue, with an 11.4% market share. The company also reported a net revenue retention rate of 139.5% in 2025.
WengAI reported revenue of RMB 250 million (USD 36.6 million) in 2023, RMB 318 million (USD 46.5 million) in 2024, and RMB 405 million (USD 59.2 million) in 2025. Over the same period, its gross margin was 44%, 50.4%, and 51.2%, respectively.
According to its prospectus, WengAI plans to build on its DI-Brain multi-agent platform and AI-for-science strategy by deepening its presence in media and communications, public services, and commercial decision-making. It also plans to expand into sectors including science and education, energy and sustainability, and healthcare. Overseas, it is targeting initial growth in Hong Kong, the Middle East, and Southeast Asia.
Net proceeds from the IPO will be used to invest in and enhance its foundation models and core R&D capabilities, expand its products and services, support brand building and customer acquisition across industries, fund potential strategic investments and acquisitions as well as overseas expansion, and meet working capital and general corporate needs.
This article was adapted based on a feature originally written by Stone Jin and published on IPO Zaozhidao. KrASIA is authorized to translate, adapt, and publish its contents.
Note: RMB figures are converted to USD at rates of RMB 6.84 = USD 1 based on estimates as of April 13, 2026, unless otherwise stated. USD conversions are presented for ease of reference and may not fully match prevailing exchange rates.