19 new unicorns valued at over 10 billion yuan emerged in half a year. Why is the primary market starting to chase "certainty"?
A striking contrast is emerging in the primary market.
On the one hand, the venture capital industry has continued to discuss the "difficulty in fundraising and exit" over the past two years; on the other hand, a small number of popular tracks such as AI large models, embodied intelligence, nuclear fusion and quantum computing have seen the resurgence of rapidly rising valuations and fiercely competed financing quotas.
According to incomplete statistics from Yicai, in the first half of 2026 alone, 19 new enterprises with a valuation exceeding 10 billion yuan were added in China's domestic embodied intelligence sector, among which the valuations of Galaxy Universal, Independent Variable, Zhi Square and Starsea Mapping have exceeded 20 billion yuan. Some enterprises have completed the jump from the 10-billion-yuan level to the 20-billion-yuan level within half a year.
The valuation of AI large models is even more staggering. Moonshot AI completed a Series F financing of over 3.5 billion US dollars in July this year, with a post-money valuation reaching 35 billion US dollars, and subsequently launched Pre-IPO round financing. The disclosed target pre-money valuation in the market has risen to 500 billion US dollars, equivalent to about 3.4 trillion yuan. At the beginning of this year, its valuation was only around 10 billion US dollars.
Capital is not flowing back evenly to the entire primary market.
What is actually happening is:
Limited active capital is rushing more fiercely into a small number of projects that seem "more certain".
1
The most scarce resource in the primary market,
is no longer projects, but exit channels
To understand this round of valuation rise, we must first figure out what the biggest pain point of the primary market has been in the past few years.
It is not that there is no good technology at all, nor that there is no capital at all.
It is the exit.
The biggest difference between primary market investment and secondary market investment is that investors cannot sell their positions at any time. When an institution invests in a startup today, the realization of returns usually relies on IPO, merger and acquisition, or old equity transfer.
As long as the exit path is not clear, no matter how high the book valuation is, it can hardly be converted into cash returns for the fund.
Therefore, when an enterprise suddenly has a clear IPO expectation, its attractiveness to investment institutions will change rapidly.
Moonshot AI is a typical case.
Public reports show that the company completed multiple rounds of financing continuously within half a year in 2026, with its valuation rising from about 10 billion US dollars all the way to 35 billion US dollars, and launched Pre-IPO financing with a pre-money valuation of 50 billion US dollars. There was even a situation where the subscription amount for financing significantly exceeded the original plan and the account was closed ahead of schedule.
What capital is buying here is no longer just "the growth of AI in the next ten years".
There is another thing:
Closer to the exit.
If an enterprise is expected to go public only after several years, investment institutions need to bear a long period of uncertainty; if it has already started share reform, listing guidance or Pre-IPO financing, even if the valuation is higher, some funds are willing to accept it instead.
This is a seemingly contradictory choice in today's primary market:
The price is higher, but institutions feel that the risk may be lower.
2
Listed companies are re-pricing
unlisted companies
The second kind of certainty comes from the secondary market.
In the past, when the primary market priced AI and robotics enterprises, it usually relied on revenue forecasts, technical teams, market space and peer financing performance.
Now a more and more important reference is the similar enterprises that have been listed.
When listed companies in the same track obtain high market value, unlisted enterprises will immediately have a new valuation anchor.
This is also why the financing heat of embodied intelligence has risen significantly in 2026. A large number of leading enterprises are promoting the capitalization process, and after companies such as Unitree Robotics have received high market attention, investment institutions will naturally recalculate the potential value of unlisted peers.
The logic is very simple.
Assuming that an unlisted robotics enterprise has a valuation of 10 billion yuan, and the market believes that its peer may reach 30 billion yuan after listing, then the 10-billion-yuan valuation still seems to have room for growth.
But the problem also arises right here.
The secondary market can reprice every day, but the primary market can hardly do that.
Once the market value of the listed company falls back from 30 billion yuan to 15 billion yuan, the "safety cushion" of the 10-billion-yuan valued company in the primary market may disappear instantly.
Therefore, the secondary market can not only push up the valuation of the primary market, but also suppress the valuation in turn.
This means that while the number of unicorns has increased significantly this year, the entire primary market is actually becoming more and more dependent on public market sentiment.
3
"The next round will be more expensive",
is becoming another kind of certainty
What is more subtle than IPO is that the transaction itself begins to create certainty.
Assuming that a company's valuation in the last round was 5 billion yuan.
The next round only raises 500 million yuan and releases very few shares, which may push the latest valuation up to 10 billion yuan.
Old shareholders thus get book floating profits, and new investors are willing to accept the 10-billion-yuan valuation because they believe the next round may reach 15 billion yuan or even 20 billion yuan.
As long as new capital keeps flowing in, this valuation system can continue to operate.
As a result, the investment logic may change.
In the past, institutions used to ask:
How much will this company be worth in five years?
Now some investors first ask:
Will someone come in at a higher price in the next round?
These two logics look similar, but their essence is completely different.
The former relies on the cash flow and profits finally created by the enterprise.
The latter relies on the transaction can continue to proceed.
This is also the most vigilant place in the primary market.
The rise in valuation itself cannot prove the increase in value.
4
A 10.6 billion yuan nuclear fusion valuation
also shows that capital is betting on "scarce tracks"
This concentration of capital has spread from AI and robotics to more early-stage hard technology sectors.
In July this year, ENN Fusion completed its first external financing, with a post-money valuation of 10.6 billion yuan after the Pre-A round. The financing funds will be used for the construction of the third-generation spherical torus hydrogen-boron fusion device, technology iteration and expansion of the R&D team.
It is worth noting that this is still a field far from large-scale commercialization.
Since 2026, at least 12 private fusion enterprises in China have announced the completion of new financing; since 2022, these enterprises have publicly completed about 24 financing rounds, 14 of which were concentrated in the first 7 months of this year.
Why are capital willing to give a valuation of 10 billion yuan when the business model has not been fully verified?
The answer, apart from the technical prospects, also lies in scarcity.
When policies, industrial trends and capital consensus all focus on one track at the same time, and the number of enterprises with mature teams and technical accumulation is limited, capital will voluntarily pay a scarcity premium.
But scarcity does not equal final success.
Nuclear fusion, quantum computing, and general-purpose robots may all have a trillion-level imagination space, but they may also take several years or even more than ten years to complete real commercial verification.
5
The primary market is shifting from
"technical risk" to "valuation risk"
This also explains why there is an important difference between this unicorn boom and the last round of Internet investment.
Today, capital does not know which embodied intelligence route will definitely win, nor does it know which large model can finally build a stable business model.
The technical uncertainty has not disappeared.
The approach capital takes is to find other quantifiable certainties:
Is there an IPO timetable;
Are leading institutions participating;
Is the next round of financing already queued;
How much will the peer be worth after listing.
As a result, the risk has actually shifted.
The technical risk has not disappeared, it has only been temporarily covered by the rising valuation.
After the valuation of an enterprise rises from 5 billion yuan to 20 billion yuan, it needs to prove more things instead.
A valuation of 5 billion yuan may only need to prove that the technology has potential.
A valuation of 20 billion yuan needs to prove that customers are willing to pay for the products.
A valuation of 50 billion yuan may need to prove that revenue can continue to expand.
After entering the 100-billion-yuan level, the market will definitely ask about profits and cash flow in the end.
The earlier the valuation overdrafts the future, the more performance the enterprise needs to achieve afterwards.
6
The real danger is not the increasing number of unicorns,
but taking "someone will take over the shares" as value
Therefore, the re-emergence of a large number of 10-billion-yuan unicorns in the primary market in 2026 does not necessarily mean a bubble, nor can it be simply understood as capital going crazy again.
AI, robotics, nuclear fusion and quantum technology do represent important future technology directions, and some of these enterprises have already generated real orders, revenue and technological breakthroughs.
What really needs to be distinguished is:
Is capital buying the long-term value of the enterprise, or the trading opportunity after the exit window is temporarily opened.
If a company can convert the financing funds into better technology, higher revenue, lower cost and stronger commercialization capability after each round of financing, then the increase in valuation has an industrial foundation.
But if the main reason for the rising valuation is only:
Some institutions have invested in the last round;
The next round will be more expensive;
The next step is to prepare for IPO.
Then the so-called "certainty" essentially means that the transaction chain has not been broken yet.
Similar stages have appeared in the past autonomous driving, computer vision and the last round of semiconductor investment boom.
When the market is at its hottest, institutions compete for financing quotas; once the financing environment changes, investors will eventually ask the same question again:
What exactly did these funds bring in return?
This is probably the most important perspective to understand the "everywhere unicorns" phenomenon in 2026.
The primary market has not suddenly become more willing to take risks.
On the contrary.
When technology and business models are getting harder to predict, it is frantically searching for certainty.
Only now the "certainty" being pursued has changed from "the enterprise will definitely make profits in the future" to:
A higher probability of going public, and a higher probability that someone is willing to pay a higher price in the next round.
These two kinds of certainty can help investment institutions shorten the waiting time, but cannot replace the enterprise to create real value.
Therefore, with 19 new 10-billion-yuan embodied intelligence unicorns added in half a year, what is really worth paying attention to is not how many companies worth 10 billion yuan have suddenly emerged in China.
But in the next stage, when capital re-requires verification of revenue, profit and cash flow:
How many of today's valuations of 10 billion yuan, 20 billion yuan or even 100 billion yuan can truly be sustained.
This is the final exam that this round of "certainty trading" in the primary market has to face.
References: Public financing information of enterprises from Yicai, Shanghai Securities News, Securities Times and *Caijing* Magazine
Risk Warning: This article analyzes the venture capital industry based on public financing and market materials, and does not constitute investment advice. The valuations of unlisted enterprises mostly come from private financing transactions with limited liquidity, and are not equivalent to the realizable value of the enterprises; part of the financing and listing plans are market-disclosed information, and the final situation shall be subject to the official announcements of the enterprises and regulatory authorities.
This article is from WeChat official account "BT Finance" (ID: btcjv1), author: Jiang Xu, authorized to be released by 36Kr.