Another specialized technology enterprise is preparing for a Hong Kong listing.

Dipu Technology, a provider of enterprise-level real-time data intelligence services, has recently resubmitted its prospectus to the Hong Kong Stock Exchange, aiming for a main board listing. This follows its initial filing in June 2024. A number of renowned institutions, including CITIC Securities, Minsheng Bank Capital, Guotai Junan International, Pudong Development Bank International, and Bank of Communications International, serve as joint sponsors.

Amidst the global wave of digital transformation, enterprise-level AI applications are becoming the cornerstone for reconstructing business logic.

As one of China's pioneers in deeply integrating AI-ready data platforms with large models, Dipu Technology is poised to become the first enterprise-level AI company to list on the Hong Kong Stock Exchange in 2025. However, its journey to listing carries both the promise of technological advancements and the realities of challenges such as competition from industry giants and an unproven profit model.

Launched by Former Huawei and Alibaba Executives: Dipu Technology's 7-Year Journey Rooted in Over 20 Years of Technological Expertise

Dipu Technology was founded in 2018 by Zhao Jiehui, a former employee of Huawei and Alibaba.

According to Tianyancha APP, Zhao Jiehui, the founder and CEO of Dipu Technology, holds a degree in Electrical Engineering and Automation from Tianjin University and boasts over 20 years of experience in the information technology industry.

Prior to starting his own venture, Zhao Jiehui spent 11 years at Huawei, where he served as a core technology expert and led the core router team. From 2015 to 2018, he worked at Alibaba Cloud, holding positions such as Senior Technical Expert and General Manager of the Enterprise Business Unit.

Given Zhao Jiehui's extensive experience, the company naturally chose to enter the enterprise services market with a "data mid-platform" to help customers integrate their disparate IT systems.

According to Zhao Jiehui, the term "mid-platform" was coined by Jack Ma. The precursor to the data mid-platform was the enterprise-level internet architecture team he assembled at Alibaba, which also contributed to Alibaba Cloud's establishment of its first batch of mid-platform technical product solutions and sales teams. At the time, Dipu Technology's core team comprised individuals with technical backgrounds from Huawei and Alibaba Cloud.

Currently, the company focuses on providing cutting-edge AI solutions for enterprises, enabling them to efficiently integrate data, decision-making, and operations on a large scale.

Especially following the recent wave of generative AI that has swept across various industries and regions worldwide, Dipu Technology has adapted to the technological shifts of the era. Leveraging its two major infrastructures—the AI-Ready FastData Foil data fusion platform and the Deepexi enterprise-level large model platform—it has implemented the deployment and implementation of Agentic AI applications in enterprises, successfully transforming into a comprehensive enterprise-level AI application solution provider.

According to Frost & Sullivan, the company is among the first in China to develop AI-ready data fusion platforms for large-model AI application solutions. Its proprietary large model is the industry's first general enterprise operation and decision-making large model to complete dual regulatory filings for deep synthesis algorithms and generative AI services.

In terms of revenue in 2024, Dipu Technology ranked fifth in China's enterprise-level large-model AI application solution market with a market share of 4.2%, expected to reach 4.4% in 2025.

Backed by Hillhouse and IDG; Dipu Faces the "Profit Challenge" in the To B Sector Despite Revenue Growth

It is worth noting that the company's commercialization process has also exhibited "stepwise growth."

From 2022 to 2024, Dipu Technology's annual revenue surged from RMB 101 million to RMB 243 million, representing a compound annual growth rate of 55.5%. Notably, revenue in 2024 increased by 88.3% year-on-year.

Behind this continuous revenue growth, the customer base has also expanded from 129 to 245. Among them, 117 major customers contribute an average annual revenue of over RMB 1.5 million, while the 33.1% repurchase rate of enterprise-level customers attests to the stickiness of its products.

Its major customers encompass leading enterprises in consumer retail, manufacturing, healthcare, transportation, and other industries. The average annual revenue per major customer increased from RMB 3.8 million in 2022 to RMB 4.8 million in 2024, indicating the substantial benefits that Dipu Technology's products bring to customers.

Meanwhile, Dipu Technology's technological edge is also evident in the optimization of its gross profit margin. From 2021 to 2023, the company's gross profit margin steadily climbed from 62.3% to 68.9%, with the proportion of high-margin businesses increasing from 51% to 74%. Notably, nearly half of this comes from the combination of data platform and large model capabilities, reflecting the robust market demand for intelligent solutions.

Funding strength serves as another indirect indicator of Dipu Technology's future growth potential.

To date, Dipu Technology has completed approximately 9 rounds of funding, totaling approximately RMB 2 billion. Notable investors include Hillhouse Capital, IDG Capital, Shanghai AI Investment Fund, Industrial Bank Asset Management, Guotai Junan, and Bank of Communications International. As early as 2021, Dipu Technology was valued at USD 1 billion, becoming a unicorn in China's data intelligence sector.

However, the flip side is that revenue growth has yet to achieve a scale effect. In 2024, the company's losses widened by 148% year-on-year to RMB 1.25 billion, accounting for 516.6% of annual revenue. Losses for 2022 and 2023 stood at RMB 655 million and RMB 503 million, respectively.

The company attributed the significant increase in losses to factors such as share-based payment expenses and changes in the fair value of shares with preferential rights.

It is difficult for the company to deny the current funding shortage. Dipu Technology stated that as it continues to invest heavily in research and development and expand its business globally, it anticipates incurring losses in the foreseeable future.

Data reveals that the company's R&D expenses in 2022 and 2023 were RMB 94.2 million and RMB 82.3 million, respectively, accounting for 93.7% and 63.8% of annual total revenue. Although R&D investment decreased to 33.5% last year, the absolute value still reached RMB 81.4 million, and the company plans to invest substantial funds in building a computing power platform in the future.

Amidst rapid expansion, Dipu Technology cannot guarantee that the company will achieve and maintain profitability in the future. In fact, high R&D investment, high losses, low revenue, and difficult self-sufficiency are common issues for AI enterprises worldwide and significant challenges faced by China's To B enterprise sector.

Enterprise AI Competition: Amidst Giant Surroundings and Vertical Competition, Dipu Technology Faces Substantial Competitive Pressure

From an industry perspective, the "data alchemy" in the blue ocean market worth hundreds of billions is gaining momentum, and Dipu Technology is poised to benefit from this.

According to Frost & Sullivan, the market size of enterprise-level large-model AI applications in China reached RMB 38.6 billion in 2024 and is projected to exceed RMB 239.4 billion by 2029, with a compound annual growth rate of 44%.

"We are not selling software; we are building the nervous system of the digital world." This sentence encapsulates the essence of the technological strategy of enterprise-level AI application service providers.

However, this also implies that giants with ecosystem strategies undoubtedly have an edge. In the realm of general large models, Dipu Technology faces ecological suppression from giants such as Baidu, Alibaba, Tencent, ByteDance, and Huawei. In vertical scenarios, competitors like the Fourth Paradigm and Transwarp Technology have established first-mover advantages.

While Dipu Technology's "industry large model + Agentic application" model presents certain barriers, the pressure for technological iteration is immense.

Furthermore, according to McKinsey research, 92% of enterprises plan to increase AI investment over the next three years, and 75% of executives believe that AI will bring disruptive changes to the industry. However, the actual implementation progress in the industry is sluggish. In 2024, only 46 enterprises (0.5% of the sample) had AI contributing more than 20% of their profits. It is anticipated that by 2025, only 1% of enterprises will reach AI maturity. This may hinder the commercialization progress of midstream and upstream service providers.

To address this, Dipu Technology is also attempting to build a service ecosystem. This includes a cloud-edge collaboration lab with Intel, reducing data processing latency to less than 5 milliseconds, and a strategic partnership with Dameng Database to achieve full-stack compatibility with domestically produced replacement solutions. More notably, its developer ecosystem layout—the ecological fission ability brought by the expansion of the DEEPNOVA open-source community—is reshaping the value chain of traditional enterprise services.

It is reported that Dipu Technology plans to utilize the funds raised from the IPO for R&D on the AI-Ready FastData Foil data fusion platform, the Deepexi enterprise-level large model platform, and the FastData and FastAGI solutions; establishing its own computing power platform; forming a team of senior experts, sales, delivery, and marketing professionals to provide specialized expansion support for the consumer retail, manufacturing, healthcare, and transportation industries; and investing in and acquiring potential targets.

Standing at the historical juncture of industrial intelligent transformation, Dipu Technology's technological breakthrough demonstrates that in a red ocean market dominated by giants, deep technological innovation remains a potent tool for breaking the deadlock.

However, the capital market is ultimately a realistic touchstone. Only by forging a replicable business closed loop can a company remain resilient in the wave of digitization. If it successfully lists this time, Dipu Technology is expected to secure a strategic buffer period of 2-3 years. In the long run, whether it can achieve sustained profitability at an early stage is crucial for truly validating the business logic of enterprise-level AI.

For this budding technology enterprise, going public is not the end but the starting point for exploring a new business civilization. Today, as data elements emerge as new means of production, it is hoped that Dipu Technology's story will have more highlights.