Under the myth of Palantir's market value of 400 billion, China's "Palantir" has emerged during the shift period of the AI industry.
Since 2024, the global capital market's focus has clearly shifted from computing infrastructure to AI applications. In the US stock market, Palantir, a big data and AI platform giant, has created a myth of a 26 - fold increase in three years, with its market value exceeding $400 billion, becoming an undisputed benchmark in this field.
Back in the domestic market, as the hype around computing power gradually fades, Chinese investors are also turning their attention to AI application layers with greater commercial potential, eagerly searching for "China's Palantir".
On October 20th, DiPu Technology, known as the "first stock in enterprise - level large - model AI applications", launched its global share offering. By 12 o'clock on the evening of the 20th, on the first day of the offering, it had recorded at least 22.11 billion yuan in margin subscription funds. Based on the public offering's fundraising scale of 35.5 million yuan, this margin subscription has achieved an over - subscription of more than 623 times, which fully demonstrates investors' high recognition of its business model and technological prospects. It's not hard to see that DiPu Technology, with its technology closed - loop and commercialization evidence driven by "Data + AI", has become a Chinese example comparable to Palantir.
DiPu Technology has built two core solutions, FastAGI (Enterprise - level Artificial Intelligence) and FastData (Enterprise - level Data Intelligence), through two technological foundations: the FastData Foil data integration platform and the Deepexi enterprise - level large - model platform. These solutions have been implemented on a large scale in multiple vertical industries such as consumer retail, manufacturing, healthcare, and transportation, creating a full - link service closed - loop for enterprise customers from multi - modal data governance to intelligent decision - making, and generating core values such as optimizing operational efficiency, improving decision - making accuracy, and enhancing productivity.
So, how should we understand the logic behind Palantir's sky - rocketing growth? Why can DiPu Technology be compared to Palantir? Where is DiPu Technology's high - growth potential?
The Secret of the Soaring US Stock AI Applications
The development of the artificial intelligence wave follows a clear value chain: from the underlying computing power and data, to the core basic models, and finally to the upper - layer applications. The value of the entire ecosystem is ultimately realized through the actual effectiveness of AI applications, which conforms to the general development law of the technology industry.
From an industrial perspective, basic model providers are also actively deploying AI applications to accelerate their commercialization process. For example, OpenAI's recent frequent actions in the hardware and software fields demonstrate its strategic intention to fully focus on AI applications.
Meanwhile, the investment logic in the capital market is also changing. In the early stage, the market was keen on investing in "shovel sellers" like NVIDIA, which provide AI infrastructure; now, the investment focus is shifting to AI application companies that can successfully achieve commercialization, the so - called "gold diggers".
It's worth noting that, different from C - end applications with unclear prospects and intense competition from giants, the B - end market has a clearer commercialization prospect, and enterprise customers are more willing to pay. This is not only because the cost - reduction and efficiency - improvement promised by AI are real commercial benefits, but more importantly, professional AI application service providers can accurately solve the four core pain points for enterprises during the implementation process:
At the data level, they can transform the enterprise's massive structured, semi - structured, and unstructured data into unified "AI - Ready" data; at the business level, relying on years of accumulated experience in vertical industries, they can achieve seamless integration of AI with actual business scenarios; at the model level, they can eliminate the uncertainty of large models and provide reliable outputs close to "zero hallucination" for business decision - making; at the security level, they can solve the data security and privacy issues throughout the process, relieving enterprises' concerns.
Software giant Palantir has become a favorite in the capital market by providing enterprises with a complete solution from the data bottom - layer to the application top - layer based on the coordinated operation of its Foundry and Artificial Intelligence Platform (AIP).
Among them, the Foundry platform forms the "semantic skeleton" and security foundation of this solution. Its core "Ontology" technology, with the help of large models, can not only automatically establish ontology objects at the data level but also update them dynamically according to the business status. By deeply associating the scattered and multi - modal data within the enterprise, it creates a dynamic business ontology that can accurately map real - world business logic, achieving real - time standardization of business at the data level.
On this basis, the AIP platform launched by Palantir in 2023 plays the role of an "intelligent front - end". Through AI assistants and low - code/no - code development tools, it effectively bridges the gap between technology and business. The combination of Foundry and AIP forms an enterprise - level "AI operating system", and its security has been strongly endorsed by providing services to top - tier enterprises and government departments.
In addition, in terms of product delivery, Palantir's "Forward Deployment Engineer" (FDE) model doesn't just dispatch technical experts. Instead, it relies on the collaboration of both business and technical key roles: the business team is responsible for clarifying customer needs and maintaining customer relationships, while the technical team quickly outputs solution prototypes and promotes implementation. This model can not only respond to customers' real - time needs and efficiently implement customized solutions but also feed back front - line business experience to the headquarters for product optimization, ensuring that the solutions are more in line with customers' actual scenarios.
According to market feedback, Palantir's growth in the US commercial market is almost entirely driven by AIP. From 2024 to 2025, a large number of customers joined the Palantir ecosystem through the "1 - Day Bootcamp" model (building an AI application prototype with AIP within one day). If the growth rate of AIP in the industry continues at the current level, it will become Palantir's main growth source from 2026 to 2027 and may even replace Foundry as the core revenue driver.
The success of this strategy is also intuitively reflected in the financial statements. From 2023 to the first half of 2025, Palantir's revenue and profit growth rates showed a remarkable acceleration, providing strong performance support for its market value in the tens of billions.
The latest view from Bank of America is that Palantir is expected to become the next winner in the AI industry wave and emphasizes that the ontology architecture and the FDE market entry strategy remain Palantir's secret weapons. Palantir's commercial sales are expected to exceed $10 billion by the end of 2030. CICC believes that Palantir's AI business logic closed - loop is accelerating.
"Data + AI": DiPu Technology Reaches the Same Goal by Different Routes
The core reason why Palantir is highly recognized in the capital market is that it provides a well - structured solution to the four pain points commonly faced by enterprises: data, integration, reliability, and security. It achieves the governance and unification of underlying data through the Foundry platform and empowers the intelligence of upper - layer business through the AIP platform, ultimately forming a closed - loop "AI operating system".
The essence of this model is to build the dual - wheel driving ability of "Data + AI". DiPu Technology, founded in 2018, has surprisingly followed a similar development path to Palantir.
DiPu Technology initially focused on enterprise data governance. Through continuous technological iteration, it developed its core business, the FastData enterprise - level data intelligence solution. In 2023, DiPu Technology innovatively built the new - generation AI - Ready FastData Foil data integration platform, which can perform word - segmentation processing on more complex enterprise data. It processes and transforms multi - modal data such as long texts, formulas, tables, and pictures into a corpus format that can be understood and processed by large models, laying the foundation for large - model training and inference.
In 2023, DiPu Technology simultaneously launched the Deepexi enterprise - level large - model platform, which supports one - stop training and inference of enterprise - specific models. In the same year, based on this technological platform, the company built the FastAGI enterprise - level artificial intelligence solution, promoting the in - depth integration of enterprise - level AI application solutions with business. This solution has been widely recognized and applied by industry customers, becoming the second growth curve for DiPu Technology's development.
If Palantir's success lies in solving the problem of AI implementation for enterprises, then DiPu Technology's FastData and FastAGI also precisely address these pain points. FastData is responsible for "data governance and standardization", while FastAGI is responsible for "intelligent transformation and business implementation". It has also derived application agents for operations, productivity, workflows, etc., acting as "high - precision, zero - hallucination" AI digital employees in vertical industries such as consumer retail, manufacturing, and healthcare.
This precise solution to the core pain points of enterprise AI implementation is also intuitively confirmed by market feedback and financial data. Financially, according to DiPu Technology's prospectus, in 2023, 2024, and the first half of 2025, DiPu Technology's operating revenues were 129 million, 243 million, and 132 million yuan respectively, with year - on - year growth rates of approximately 28.4%, 88.3%, and 118.4% respectively, showing a similar trend of continuously increasing performance growth to Palantir.
What's even more noteworthy is the change in its business structure. Looking at different product lines, the revenue of the FastAGI enterprise - level artificial intelligence solution has increased sharply since its launch in 2023, growing from 6.55 million yuan in 2023 to 90 million yuan in 2024. As of the first half of 2025, the revenue of this business was 73 million yuan, a year - on - year increase of 192%, significantly higher than the 68.57% revenue growth rate of the FastData enterprise - level data intelligence solution during the same period. This has also pushed the proportion of the FastAGI enterprise - level artificial intelligence solution in DiPu Technology's total revenue from 5.1% in 2023 to 55.3% in the first half of 2025, becoming the company's new growth curve.
Figure: Composition of DiPu Technology's Operating Revenue. Source: Wind, 36Kr compilation
Looking at its downstream customer composition, DiPu Technology's customers are widely distributed in various fields such as consumer retail, manufacturing, healthcare, and transportation. In terms of customer expansion strategy, DiPu Technology adopts the "lighthouse customer radiation" model. It first accumulates vertical industry implementation capabilities through cooperation with leading enterprises in the industry and then attracts new customers in the same industry.
As of December 31, 2022, 2023, and 2024, and June 30, 2025, DiPu Technology had provided services to 129, 178, 245, and 283 customers respectively. Since the switching cost for industry customers after cooperating with an AI application solution provider is extremely high, AI application solution providers will accumulate more and more existing customers as the business develops. Therefore, DiPu Technology's revenue scale also has a "compound interest effect".
Overall, whether from the perspective of technological foundation, business layout, or financial growth, DiPu Technology shows similarities to Palantir in many aspects, which is the core reason why the company's IPO has attracted so much attention.
Figure: Industry Composition of DiPu Technology's Downstream Customers. Source: Wind, 36Kr compilation
Imagination Space and Valuation Premium
Looking back at Palantir's sky - rocketing growth, it is driven by rapid performance growth and continuous increase in valuation.
Since 2023, Palantir's stock price has increased 26 times, corresponding to an approximately 100% increase in operating revenue, and its PS has grown from about 10 times to over 100 times. This clearly shows that once an AI application company verifies the effectiveness and high - growth potential of its business model, the capital market is willing to give it a valuation premium far higher than that of traditional software companies.
So, what is the imagination space for DiPu Technology's performance?
First, looking at the industry space, in recent years, China has also significantly increased its attention to AI applications, promoting the integration of AI with various industries, which has become a top - level design.
In the "Three - Year Action Plan for 'Data Elements ×' (2024 - 2026)", the document clearly states the need to improve the quality of data supply, especially to meet the high - quality data sets required for large - model training, and to promote applications in multiple industries such as industrial manufacturing, modern agriculture, commercial circulation, and transportation.
In August 2025, the State Council issued an opinion on in - depth implementation of the "Artificial Intelligence +" action. The opinion points out the need to promote the extensive and in - depth integration of artificial intelligence with all industries and fields of the economy and society. In terms of policy goals, by 2027, artificial intelligence will be widely and deeply integrated with six key fields, and the penetration rate of applications such as new - generation intelligent terminals and agents will exceed 70%; by 2030, the penetration rate of these applications will exceed 90%, and the intelligent economy will become an important growth pole for China's economic development.
Frost & Sullivan points out that from 2024 to 2029, the market for enterprise - level artificial intelligence application solutions in China will grow from 38.6 billion to 239.4 billion yuan, with a compound growth rate of up to 44%. In the more segmented market for enterprise - level large - model artificial intelligence application solutions, the corresponding industry space will grow from 5.