Hong Kong Stocks' New AI Narrative: Embracing "Token Economics"

In early spring 2026, Hong Kong stocks’ AI sector is undergoing a valuation logic overhaul, becoming the focus of global capital. As of the noon session on March 24, the Hang Seng Tech Index was at 4,779.52 points, up 1.42%. Notably, the Hang Seng Hong Kong Stock Connect AI Thematic Index, which reflects the overall performance of the AI industry, has risen by about 15% year to date, showing that capital is accelerating its concentration into the AI track.

In this wave, Hong Kong-listed China-made AI “twin stars” of large models have been especially eye-catching. At the noon session on March 24, MiniMax’s latest share price was HK$999, up 9%, with a total market cap of approximately HK$313.32B; while Zhipu AI’s latest share price reached HK$627, up 6.27%, with a total market cap of HK$279.54B. Since listing, the year-to-date share price gains of both companies have exceeded 400%.

Behind the market’s热 pursuit, a clear valuation main line is coming into view: when AI agents such as OpenClaw ignite the “application rollout” wave, Token consumption driven by single tasks is increasing exponentially. As the smallest unit through which AI large models process information, Token is becoming the core pricing unit for measuring both the depth of AI business rollout and commercial potential.

In this competitive landscape shift triggered by the AI large-model wave, Hong Kong investors and analysts are also重新 seeking the “ruler” for measuring value.

Reshaped Competitive Landscape

The change in the valuation anchor for Hong Kong stocks’ AI sector begins with shifts in the competitive landscape of the Hong Kong stock market.

Market analysis believes that the current Hang Seng Tech Index still has traditional internet and software companies as its core weighting. But in this AI revolution, large models are becoming the new traffic entry point and value center of gravity, while traditional platforms and software are having their user time and commercial space continuously eroded.

In January 2026, MiniMax and Zhipu AI became among the first global listed companies for large-model R&D. After the share prices of China-made AI large-model “twin stars” rose by several multiples, their market caps have already come level with Baidu and are comparable to JD.com.

“The listing of technology companies such as MiniMax and Zhipu AI could allow investors to directly participate in China’s AI development, while also freeing them from the constraints of being passively exposed to traditional businesses like e-commerce and food delivery when investing in the so-called Big Eight technology giants.” Chen Mingkang, a senior equity strategy analyst at Bloomberg Intelligence, told a reporter. “The large language models these two companies are developing rank among the top globally, and their products have demonstrated strength in both performance and cost efficiency.”

Chen Mingkang believes that since listing, the China-made AI large-model “twin stars” have seen their share prices rise by several multiples, which may reflect optimistic expectations from the market regarding prospects for early adoption and demand for high-performance, cost-efficient models.

The market expects that the year-over-year revenue growth rates of MiniMax and Zhipu AI will both exceed 150%, far higher than the single-digit high-growth levels of China’s Big Eight technology giants, as well as the low double-digit growth levels of the U.S.’ Big Seven technology giants.

If the story of the new wave of AI is “from 0 to 1,” then what traditional internet giants are playing out is a leap of “from 1 to 10.”

Since this year, capital expenditure plans by old-guard tech giants such as Tencent, Alibaba, and Baidu have frequently drawn market attention. To compete for AI entry points, these players with strong cash flows are not hesitating to pour large sums into compute infrastructure and model R&D.

“Old players are running into new headwinds—if they increase capital expenditures, it must mean they’re confident in the development of the sector. Using new business models to pair with existing business models to achieve geometric growth is the pattern that traditional internet giants will either choose proactively or feel compelled to adopt.” Wang Xinjie, chief investment strategist at Standard Chartered China Wealth Solutions Department, pointed out. “In the current business models across the industry, all kinds of application scenarios will definitely emerge. For internet giants, increasing capital expenditures and then monetizing quickly is the direction for development right now.”

Chen Mingkang predicts that cloud service giants control full-stack AI infrastructure and may also roll out bundling strategies, creating advantages in both pricing and profit margins. As competition intensifies, industry rivalry may further escalate.

As AI agents such as OpenClaw ignite the “application rollout” wave, internet giants are accelerating their shift from “model competitions” to “securing seats in scenarios.” Tencent has quickly launched WorkBuddy, an all-scenario AI agent compatible with OpenClaw skills; Alibaba Cloud has launched a dedicated image service; and JD Cloud is running pilot programs in business lines such as retail and logistics. These old players with massive user ecosystem are trying to open up new growth space through OpenClaw.

This collective push to embrace OpenClaw quickly produced feedback in capital markets. Tencent became one of the biggest winners: just one day after WorkBuddy launched on March 10, its share price surged at market open, reaching a peak of HK$556 during the session, and ended the day up 7.27%—the largest single-day gain in nearly three months. Trading value exceeded HK$30 billion, and market expectations for AI commercial rollout were fully ignited. Alibaba, whose cloud arm is Alibaba Cloud, did not experience a one-day blowout, but benefited from Bailian MaaS, which saw a record-high growth rate due to a surge in Token consumption; its share price gradually rebounded during the technology stock rebound wave in mid-March, confirming the market’s recognition of its Token business strategy. After the release of news about pilots in retail and logistics scenarios, JD.com Group also broke away from the previous sideways trading trend. From March 12 to 16, it closed higher for five consecutive trading days, with Hong Kong stock prices stepping up from HK$108.6 to HK$111.5.

When OpenClaw pushes AI from “Q&A interaction” to “task execution,” Token consumption per single task rises exponentially. Token is no longer just a unit of compute measurement at the technology level—it has become the core pricing unit for measuring the depth of AI business rollout and the potential for commercialization. This is the key logic starting point behind the valuation anchor shifting toward Token.

Hard Currency in the Digital Economy Era

In March 2026, at NVIDIA’s GTC conference, the company’s founder and CEO Huang Renxun introduced a new concept to reshape industry understanding—“Token factory.” In his view, data centers are undergoing a fundamental role transformation: they are no longer the “electronic warehouses” used in the past to store files and data, but an intelligent production line running day and night. This factory takes in electricity and data and produces Token.

As Huang Renxun said, Token has become the “hard currency” of the digital economy era, and its generation efficiency will directly determine the survivability of technology companies and their revenue curves.

What is Token?

In simple terms, Token is the smallest unit for every interaction between a user and a model: every word and every punctuation you input to the AI, as well as every character the model replies with, will be broken down and consumed in a certain number of Tokens.

For investors, Token has become a focal point because for the first time it unifies the revenue side and cost side of AI companies into the same measurable scale.

“For the AI sector, traditional PE valuation models don’t work very well. For these potentially high-growth industries, investors mainly focus on their future potential profitability.” Wang Xinjie told a reporter.

Wang Xinjie pointed out that at the current deployment stage, large capital expenditures may make it difficult for these companies to obtain profits immediately. Therefore, in the AI industry, current user traffic and Token consumption are what investors pay more attention to, because they can increase the probability of future profitability.

For example, when analysts issue valuation reports for AI companies, they start not only looking at traditional income statements, but also trying to estimate each company’s share of Token consumption in the global large-model subscription and API markets, as well as the gross profit that each unit of Token can generate.

Behind this shift in valuation logic lies the market’s renewed understanding of the essence of AI companies’ business models: growth in Token consumption means higher user activity and deeper business penetration; while improvement in monetization ability per Token indicates that the business model is moving toward maturity.

More specifically, native large-model companies win high valuations thanks to their technical foundation, because investors believe their Token consumption can achieve exponential growth. Meanwhile, traditional platform giants with massive scenario coverage are given higher monetization imagination, because existing businesses such as e-commerce recommendations, food delivery dispatch, and content distribution naturally provide channels to convert Token calls into real revenue.

Yang Delong, chief economist at Qianhai Open-Source Fund, told a reporter that in the AI era, Hong Kong stocks place more emphasis on new indicators such as customer traffic. This change in valuation logic will also have a major impact on how technology stocks in Hong Kong are priced. Investors will value technology companies’ advantages in Token and big data more.

When Token Becomes the New Unit of Account

Under this trend, Token is becoming a new valuation anchor for AI companies. Song Weiwei, manager of the CSI China-Hong Kong Robotics Index Fund, told a reporter that OpenClaw represents a paradigm shift for AI from “conversation” to “execution,” and Token consumption is the first principle behind the entire investment chain.

The business model of pure AI companies relies highly on Token consumption, and the underlying logic for benefiting from the direction is the transmission of this multiplier effect.

The business data of Hong Kong AI “twin stars” is the best proof. Driven by the explosive growth of its M2 series text models, MiniMax’s daily Token consumption surged by more than 6 times compared with December 2025, driving 2025 full-year revenue to skyrocket 158.9% year over year to US$79.04M. Its gross margin also jumped sharply from 12.2% to 25.4%, demonstrating outstanding commercialization efficiency.

Zhipu, on the other hand, has relied on the rapid commercialization of the GLM-5 flagship model API service, with Token consumption continuing to climb. In November 2025, average daily Token consumption reached 42 trillion, driving its 2025 revenue to grow by multiples. Its MaaS platform has brought together over 3 million enterprise and application developers, and API pricing was cumulatively raised by 83% in the first quarter of 2026, with a clear trend of both volume and price increasing.

The common feature of these two companies is that Token call volume has become a direct revenue indicator, and performance delivery pushes the valuation from being “concept-driven” to being “revenue-driven.”

Duannairong, senior investment strategist at RBC Wealth Management, told a reporter that currently most “new forces in AI” have not yet turned profitable, so the market pays more attention to their potential for dynamic growth.

Duannairong believes that in the context of “Token inflation,” the value anchor for Hong Kong technology stocks may be referenced through the following dimensions. First is the conversion efficiency from Token consumption to revenue—that is, how much revenue each unit of Token can generate—which determines the true profitability of the business. Second is the monetization rate of traffic and customer stickiness, including metrics such as customer retention and paid renewal rates, which reflect the sustainability of revenue. Third is the locking capacity of prepaid payment systems—how much future revenue companies can lock in through prepayment, subscriptions, and so on—better reflecting the stability of the business.

It is also worth noting that the performance and cost-effectiveness advantages of domestically made AI models are supporting their Token competitiveness, and Token pricing and cost advantages are also one of the foundations for building valuation premiums.

Data on OpenRouter, a third-party API aggregation platform, shows that in late February 2026, API Token usage using China-made AI models surpassed that of U.S. AI models for the first time.

A recent report from Goldman Sachs obtained by the reporter points out that the performance gap between China’s new-generation AI models and U.S. models has been significantly narrowing. It supports ultra-long context windows of up to 1 million Tokens and powerful agent capabilities. Second, the Token pricing for China’s AI models is only 5% to 10% of that of the U.S. flagship models, making the value-for-money even more prominent. Finally, China’s AI models perform exceptionally well in multimodal fields such as text, video, and audio, further expanding Token application scenarios.

In its report, Goldman Sachs analyzes that the surge in Token demand will bring revenue breakthroughs for Chinese AI model companies while also推动 cloud service revenue growth.

In terms of the competitive landscape, independent AI model companies are expected to rise quickly by leveraging their Token business, competing with internet giants. Top internet companies also will consolidate their advantages in Token-related infrastructure through capital expenditure investments.

Bloomberg Intelligence’s industry research report notes that explosive share price rallies by companies such as MiniMax and Zhipu AI are, in essence, the market’s recognition of the valuation of their combination of “low token cost + high demand growth rate,” which enables their market caps to quickly come to parity with technology giants such as JD.com and Baidu.

For investors, it may be necessary to accept a hard reality: in an AI industry that evolves rapidly, perhaps there is no permanent valuation anchor. Today’s Token may be replaced tomorrow by a new unit of account. Today’s leaders may be overturned tomorrow by cross-industry entrants.

What can be confirmed is that Hong Kong stocks are becoming the core stage for docking China’s AI industry with global capital. On this stage, the story has only just begun.

(Zhang Weize also contributed to the interview for this article.)

(Author: Yuan Sijie; Editor: Zhu Lina)

Huge volumes of information and precise analysis—on the Sina Finance app

责任编辑：江钰涵