New Stock News | DataCanvas Files with the Hong Kong Stock Exchange, Aiming to Become the First AI Computing New Cloud Stock

According to a disclosure by the Hong Kong Stock Exchange on September 29, Beijing DataCanvas Technology Co., Ltd. (referred to as: DataCanvas) has submitted a listing application to the Main Board of the Hong Kong Stock Exchange, with China Galaxy International acting as its sole sponsor. According to the prospectus, DataCanvas is a leading AI computing service provider in China, purpose-built to support compute-intensive AI workloads spanning large-scale model training and inference. The company builds, operates, and monetizes AI computing power. It provides one-stop AI computing infrastructure solutions for AI data center ("AIDC") projects and offers on-demand cloud computing services driven by its proprietary computing operating system. According to CIC data, measured by 2025 revenue, DataCanvas ranks ninth among AI computing service providers in China and is the only AI computing new cloud ("AI computing new cloud") among the top ten, making it the largest AI computing new cloud in China. As an AI-native infrastructure provider, DataCanvas differentiates itself from hyperscale AI computing service providers by building its architecture from the ground up around high-density computing deployment, heterogeneous resource orchestration, and the operational stability required for model-centric computing demands. Driven by generative AI (spanning large-scale model training and high-concurrency inference), the entire computing industry is undergoing a fundamental transformation. General-purpose cloud architectures were previously designed for transactional and scaled web workloads. These composite workloads have given rise to a distinctly different set of infrastructure requirements that traditional CPU-centric cloud environments cannot satisfy. Therefore, according to CIC data, compute-optimized AI-native infrastructure is essential to sustaining continued AI progress. DataCanvas established its capabilities through AI foundational software ("AIFS") services, and subsequently in early 2023 strategically concentrated its resources on AI computing infrastructure and cloud services. As enterprise adoption began to accelerate, the company extended its model engineering expertise in data processing, training orchestration, and deployment to the infrastructure layer. DataCanvas lowers the barrier to entry for AI computing through heterogeneous scheduling, serverless orchestration, and usage-based billing. The company's proprietary "One Degree Computing Power" standard measures computing power in standardized computing power units ("DCU") and aligns billing with actual computing power consumption (rather than reserved physical capacity). According to CIC data, this is the first such quantifiable billing unit in China's AI computing industry. DataCanvas has also begun extending this concept to model inference, packaging computing clusters and selected third-party foundation models into standardized inference outputs that customers access through the company's application programming interface ("API") and pay for based on token consumption. DataCanvas's business has achieved rapid expansion at both the operational and financial levels. To ensure operational consistency, the company exercises comprehensive control over its computing resources throughout their entire operational lifecycle, covering facility architecture planning, system deployment, and daily dynamic scheduling driven by the company's proprietary computing operating system. As of the Latest Practicable Date, under its "central scheduling + regional nodes" framework, DataCanvas's controllable computing power in China has exceeded 29,000 PFLOPS across seven provinces, which, according to CIC data, positions DataCanvas among China's leading AI computing new clouds by contracted lifecycle-managed capacity. As of the Latest Practicable Date, DataCanvas has served over 150 enterprise customers through AI computing infrastructure solutions and AI computing cloud services. On the financial side, DataCanvas's revenue grew from RMB 106 million in 2023 to RMB 396 million in 2024 and RMB 1,098 million in 2025, and from RMB 306 million for the six months ended June 30, 2025 to RMB 588 million for the six months ended June 30, 2026, representing nearly a tenfold increase over two years, driven by accelerating enterprise demand for intelligent infrastructure and deepening commercialization across both business lines.