Analysis of the Impact of Zhipu AI's Commercialization Layout on AI Unicorn Valuation Logic
Unlock More Features
Login to access AI-powered analysis, deep research reports and more advanced features
About us: Ginlix AI is the AI Investment Copilot powered by real data, bridging advanced AI with professional financial databases to provide verifiable, truth-based answers. Please use the chat box below to ask any financial question.
Based on the latest market data and industry dynamics, I will systematically analyze the impact of Zhipu AI’s accelerated commercialization layout on the valuation logic of unprofitable Chinese AI unicorns for you.
Zhipu AI was officially listed on the Hong Kong Stock Exchange on January 8, 2026, becoming the world’s first listed company with general artificial intelligence (AGI) foundation model as its core business [1]. The IPO offering price was set at HK$116.20 per share, and it closed at HK$131.5 on the first day of trading, representing a 13.17% increase, with a total market capitalization reaching HK$57.89 billion. The Hong Kong public offering was oversubscribed 1,159.46 times, and the international offering was oversubscribed 15.28 times, showing high recognition from the capital market [1].
Meanwhile, on January 5, 2025, Zhipu AI established a new company in Xiamen - Xiamen Zhipu Huazhang Technology Co., Ltd., with a registered capital of RMB 5 million, whose business scope includes artificial intelligence application software development, artificial intelligence theory and algorithm software development, etc., and it is incorporated into the second category of value-added telecommunications services [2]. This layout marks a strategic upgrade of its commercial operation capabilities.
Zhipu AI has formed a standardized product system centered on MaaS (Model as a Service). Its full-year 2024 revenue reached RMB 312.4 million, with a year-on-year growth of 130%; its first-half 2025 revenue was RMB 191 million, with a year-on-year growth rate of 325% [3]. From the perspective of revenue structure:
| Business Type | 2024 Revenue | Percentage | H1 2025 Percentage |
|---|---|---|---|
| Local Deployment | RMB 263.9 million | 84.5% | 84.8% |
| Cloud Deployment (MaaS) | RMB 48.48 million | 15.5% | 15.2% |
Zhipu AI plans to increase the revenue share of its API business to 50%, and expects its total revenue to achieve a year-on-year growth of over 100% in 2025 [4]. As of the first half of 2025, Zhipu has provided large model services to more than 8,000 institutional customers, and its MaaS platform has gathered over 2.7 million enterprise and developer customers [5].
Zhipu AI’s listing marks a shift in the narrative logic of AI large model vendors from “telling technology stories” to “commercial value realization” [6]. Market valuation has shifted from traditional user growth expectations and technological barrier assessment to greater emphasis on actual revenue scale, profitability paths, and cash flow sustainability.
- From user valuation to revenue valuation: Investors pay more attention to ARR (Annual Recurring Revenue) and revenue growth rate
- From PS valuation to composite valuation: PS (Price-to-Sales ratio) is used for pricing, and Zhipu’s expected PS ratio for 2026 is approximately 30x [7]
- From technical indicators to commercial indicators: Developer ecosystem and government-enterprise implementation capabilities have become core assessment dimensions
The prospectuses of Zhipu AI and MiniMax reveal two completely different commercialization paths, resulting in significant valuation differences:
| Dimension | Zhipu AI | MiniMax |
|---|---|---|
| Commercial Path | B-end local deployment + API | C-end overseas expansion + multimodal applications |
| Gross Profit Margin | 56.3% | 12.2% |
| 2024 Growth Rate | 130% | 782.2% |
| Offering PS | ~30x | Relatively higher (dynamic PS is lower) |
Huaan Securities pointed out that this differentiation indicates that the AI large model industry is entering a brutal elimination stage [8]. Different commercial paths correspond to different valuation logics:
- B-end deep cultivation type: Emphasizes customer stickiness and customized service capabilities, adopting a higher PS valuation
- C-end overseas expansion type: Emphasizes user scale and growth flexibility, but faces challenges in monetization efficiency
Zhipu AI’s commercial practice has established a valuation reference for the entire industry, triggering the following key changes:
Prospectus data shows that Zhipu AI consumed an average of approximately RMB 221 million in operating funds per month in the first half of 2025 [9]. Investors require enterprises to clearly disclose the survival period that cash reserves can support, which directly affects the safety margin of valuation.
The gap between Zhipu AI’s 56.3% gross profit margin and MiniMax’s 12.2% gross profit margin is significant [9]. High gross profit margin means stronger risk resistance and valuation premium space, and the quality of the business model is replacing pure scale expansion as the core of valuation.
Zhipu AI’s revenue mainly comes from the government, large state-owned enterprises, and leading internet companies. 9 out of the top 10 internet companies are using its GLM large model [10]. The stability and renewal rate of B-end customers have become important support for valuation.
Zhipu AI’s Southeast Asian revenue accounts for 11.1% of its local deployment revenue, and it has significant growth potential in overseas markets [11]. Internationalization capability has become a key word to open up the valuation ceiling.
Chen Shi, Investment Partner of Frees Fund, pointed out: “Whether we can run through commercial implementation within the time gained is the watershed that determines its final valuation level.” [12] This means that technological leadership must be converted into sustainable commercial revenue.
Zhipu’s successful listing provides a valuation anchor for entities such as Moonlight AI (Yuezhi Anmian) and StepUp AI (Jieyue Xingchen). These enterprises need to quickly deliver replicable industry large orders and narrow losses to keep up with the listing rhythm in 2026 [8].
Enterprises such as MineLLM (Mianbi Zhineng) and 01.AI (Lingyi Wanwu) have been forced to cut down or reduce pre-training, and shift to industry Agent and scenario kit applications. The valuation logic of such enterprises has shifted from “foundation model” to “solution provider”, and the PS valuation multiple will be significantly compressed [8].
Zhipu AI’s listing has verified the applicability of PS valuation in Chinese AI unicorns. Referring to the comparison between Kingsoft Corporation (2024 revenue of RMB 10.3 billion, market capitalization of approximately RMB 39.6 billion) and OpenAI (first-half 2025 revenue of US$4.3 billion, valuation as high as US$500 billion) [12], Chinese AI unicorns need to find a reasonable valuation position from a global perspective.
Zhipu AI’s accelerated commercialization layout is profoundly reshaping the valuation logic of Chinese AI unicorns. The core conclusions are as follows:
- Valuation Paradigm Shift: From “technology storytelling” to “commercial value realization”, with revenue scale, gross profit margin, and cash flow becoming new valuation anchors
- Intensified Differentiation: Different commercial paths correspond to different valuation logics, with both B-end deep cultivation and C-end overseas expansion having their own valuation rationality
- PS Valuation Becomes Mainstream: Zhipu’s expected PS ratio of approximately 30x for 2026 provides a reference benchmark for the industry
- Narrowing Window of Opportunity: 2025 remains the last listing window for the “Six Little Dragons”. If Zhipu and MiniMax successfully list, they will absorb most of the overseas liquidity [8]
- Long-Term Value Anchoring: Whether Zhipu can continue to run through commercial implementation after listing will determine the long-term valuation level of the entire AI unicorn sector
Zhipu AI’s listing is not only a testament to the technical strength of the Tsinghua system, but also a watershed in the valuation logic reconstruction of the entire AI industry chain. In this transformation, enterprises that can convert technological advantages into sustainable commercial revenue will gain long-term favor from capital.
[1] Guancha.cn - “First Large Model Stock” Zhipu Rises 13% on Debut, Company to Launch Next-Generation Model (https://www.guancha.cn/economy/2026_01_08_803199.shtml)
[2] CN Energy News - Zhipu AI Establishes New Company in Xiamen with Registered Capital of RMB 5 Million (https://www.cnenergynews.cn/article/4PqMbRGCTqx)
[3] PE Daily - Zhipu AI Officially Lists Today, 6 Things You Need to Know (https://news.pedaily.cn/202601/559655.shtml)
[4] Securities Times - Beijing “Unicorn” Launches IPO, Expected to Become Global “First Large Model Stock” (https://www.stcn.com/article/detail/3545998.html)
[5] Sina Finance - Overseas Expansion and Listing: China’s First Batch of Large Models Have Made It (https://finance.sina.com.cn/tech/roll/2026-01-04/doc-inhfcasw7648852.shtml)
[6] Huaan Securities - Analysis of Valuation Logic for AI Large Model Vendors (https://finance.sina.com.cn/)
[7] Wall Street CN - “First Domestic Large Model Stock” Zhipu Rises Narrowly on Debut: Model Iteration × Ecological Flywheel (https://wallstreetcn.com/articles/3762845)
[8] 21st Century Business Herald - “First Large Model Stock” Starts, Zhipu and MiniMax Bet on Different Paths (https://stcn.com/article/detail/3547970.html)
[9] 36Kr - Dissecting the Prospectus, Understanding Two “Lifestyles” of Chinese Large Model Unicorns (https://m.36kr.com/p/3608733082223874)
[10] Securities Times - “First Domestic Large Model Stock” Zhipu Rises Narrowly on Debut: Model Iteration × Ecological Flywheel (https://stcn.com/article/detail/3547970.html)
[11] PE Daily - Zhipu AI Officially Lists, Full Analysis of the Most Beneficial Companies (https://www.popcj.com/depth/2739482601447954)
[12] Tencent Technology - Dissecting the Prospectus, Understanding Two “Lifestyles” of Chinese Large Model Unicorns (https://m.36kr.com/p/3608733082223874)
Insights are generated using AI models and historical data for informational purposes only. They do not constitute investment advice or recommendations. Past performance is not indicative of future results.
About us: Ginlix AI is the AI Investment Copilot powered by real data, bridging advanced AI with professional financial databases to provide verifiable, truth-based answers. Please use the chat box below to ask any financial question.