Kimi and Its Peers Are Rewriting the Rules of the Valuation Game

On June 8th, Bloomberg reported that Moonshot AI is seeking a new round of financing, planning to raise up to $2 billion, with a post - investment valuation target of $30 billion.

In RMB, it's approximately 200 billion yuan.

What scale is this? It's about twice the market value of Haidilao and higher than most A - share listed companies. And this valuation is created by an AI company named Kimi, which has only been established for three years, has a total of more than 300 employees.

$30 billion is not even the highest in this field. During the same period, Zhipu has been listed on the Hong Kong Stock Exchange, with its market value once exceeding $90 billion; DeepSeek is reported to be in talks for its first round of external financing, with a valuation of about $45 billion; even MiniMax, which was founded around the same time as Kimi, has landed on the Hong Kong Stock Exchange with a valuation of $6.4 billion.

The valuations of Chinese large - model companies have skyrocketed in the past six months.

How did these numbers come about? What's going on behind them?

01 The End of an Old Logic

To understand this collective price increase, we need to go back two years.

In 2024, the valuation anchor for large - model companies was the monthly active user count. At that time, Kimi was the best at playing the traffic game. Relying on the differentiated selling point of "2 million - word super - long text", its monthly active users once reached 36 million, making it the top - notch large - model at that time. The industry generally believed that the more users a company had, the more promising its future would be.

This is the valuation inertia left over from the Internet era. In the past two decades, every financing round of BAT companies has told the same story: attract users first and then talk about monetization. Investors have been trained by this logic for an entire generation.

But large models are not apps.

The strategy of burning money to gain traffic has encountered a fatal problem in this field: inference cost. Every time a user calls a large model, it consumes real computing power. The marginal cost does not approach zero as the number of users increases; instead, it rises. ByteDance, Alibaba, and Tencent can make their products free with almost unlimited subsidies, but ordinary startups simply can't keep up.

By March 2026, Kimi's monthly active users had dropped to 8.34 million, ranking behind Doubao (345 million), Qianwen (166 million), and DeepSeek (127 million), and it couldn't even make it into the top five. According to the Internet valuation logic, Kimi should have been out long ago.

But its valuation actually increased from $4.3 billion to $30 billion during this period.

What happened in between? The answer is that the rules of the valuation game have been quietly rewritten.

Investors no longer ask "how many users do you have", but start to ask "how deep is your technological barrier", and a more direct question: "how much money are you making now, and what's the growth rate?"

The answers given by Kimi made the market take a second look at it.

In January 2026, Kimi's paid orders increased by 8280% month - on - month, more than eighty times. Its Stripe payment ranking jumped from outside the top 100 to the 22nd in the world, and a month later, it rose to the 9th. By March, the company's ARR (Annual Recurring Revenue) exceeded $100 million; in April, it exceeded $200 million. This speed is top - notch among global SaaS companies.

What's more noteworthy is where the money comes from. Kimi's revenue explosion mainly comes from two aspects: one is paid subscriptions, and the other is B - side API calls. The latter means that more and more developers and enterprises are using Kimi's model capabilities to build their own products. The well - known code tool Cursor was reported to have called Kimi's model at the underlying level. By the way, Cursor's own valuation is also as high as $50 billion.

This is a new valuation logic: the value of large - model technology as infrastructure. It is no longer compared to Internet apps but is more similar to semiconductors or EDA software. The number of users may not be the largest, but every user depends on it and is willing to pay real money for it.

02 The Darkest Hour and Strategic Refocus

But Kimi's journey to the present has not been smooth. In fact, just before this commercial explosion, it was in the most difficult time.

At the beginning of 2025, DeepSeek emerged out of nowhere. With its low cost and high performance, it quickly became a hot topic at home and abroad, becoming a phenomenon in the AI circle. For Kimi, the impact of this event came not only from external competition but also from within. During that period, an employee said directly at the annual meeting: "In the current situation, why would job - seekers come to our Kimi instead of DeepSeek?"

Kimi's response was a complete strategic refocus.

In terms of direction, it cut off the exploratory projects that scattered its energy and fully bet on "model first". Its technical route shifted from "long - text assistant" to "Agent model for complex tasks", focusing on three dimensions: Token efficiency, long context, and Agent Swarms. In terms of products, Kimi K2.5 was released at the beginning of 2026, and its call volume on the OpenRouter platform immediately ranked first, directly triggering that round of paid growth.

Interestingly, the first reaction of Kimi's algorithm team to DeepSeek's popularity was not anxiety but "excitement". A saying circulated within the company is that "DeepSeek saved us" because it proved the direction of the technical route and forced the team to find its own position.

This mindset may also explain the nature of Kimi. With more than 300 employees, the average age is less than 30, and 80% have introverted personalities. The company has no departments, no job ranks, and no OKR. In 2025, a 17 - year - old high - school intern published a paper as the first author, which was then reposted and evaluated by many well - known figures in Silicon Valley. This also indirectly proves Kimi's corporate culture: as long as the direction is right, age and seniority don't matter.

03 A Collective Re - evaluation

Kimi's story is of course not an isolated case.

If we look at the period from the end of 2025 to the first half of 2026, we'll find something interesting: the valuations of almost all top - tier large - model companies soared intensively during this period. Zhipu completed its listing on the Hong Kong Stock Exchange and has launched an A - share IPO; MiniMax also landed on the Hong Kong Stock Exchange; DeepSeek is reported to be in talks with institutions such as the National Integrated Circuit Fund for its first round of external financing; Jieyue Xingchen is also rumored to be about to submit an application for listing on the Hong Kong Stock Exchange.

In six months, the valuation coordinate system of this field has been raised by an order of magnitude, and the entire industry has reached the same critical point.

This critical point can be described by a set of numbers: in 2026, the monthly active user scale of Chinese AI - native apps reached 440 million, with a single - quarter increase of 130 million (data from QuestMobile). This means that large - model products are changing from "hot words in tech media" to daily tools for ordinary users, and the willingness to pay is starting to rise. Kimi's ARR rising from zero to $200 million within a few months is a clear industry signal, telling the market that "large models can really make money".

This is a bit like the mobile Internet in 2013 and 2014. In those two years, the trigger for the market's re - pricing was the emergence of the first batch of native apps that really made money. WeChat started its commercialization, and Didi developed a unit economic model. Once someone proves that "there is really money to be made on this path", the entire valuation system starts to be rewritten. The trigger for the large - model field today is the same mechanism, but this time, the speed is faster and the numbers are larger.

Of course, there are also aspects worth considering in this re - evaluation.

Taking Kimi as an example again, an ARR of $200 million is not a small amount, but corresponding to a target valuation of $30 billion, the PS multiple is about 150 times. This number is not justifiable in any mature industry. Even in the high - growth SaaS field, the PS multiples of top - tier companies are usually between 30 and 50 times.

On the other hand, the pressure of burning money is real. Data shows that MiniMax's monthly cash consumption is about $27.9 million, and Kimi's scale and investment are no less. With a cash reserve of 10 billion yuan on the books, there is still pressure in the face of continuous model training and computing power investment.

But the other side of the logic that supports these valuations also exists. OpenAI's valuation reached $300 billion in 2025, and Anthropic's valuation in June 2026 is said to have exceeded OpenAI, reaching $965 billion. If Chinese model companies can continue to maintain their competitiveness in technology and the market scale they serve is not inferior to any economy, whether $30 billion is high or low is itself a question without a standard answer.

Valuation is ultimately an expectation. It measures not today's revenue but the market's collective judgment of where a company can reach at a certain future point.

Conclusion

On the day when the news of Kimi's $3 - billion financing came out, neither Kimi nor Yang Zhilin made any response.

In the same month, at the end of his speech at the Tsinghua AGI Summit, he shared a conversation between himself and Kimi. He asked Kimi: "The arrival of AGI may threaten humanity. As a researcher, should we continue?" Kimi's answer was: "Even if there are risks, we will still continue, because giving up AGI means giving up the potential of human civilization." Yang Zhilin agreed with that answer and then stated his plan: "I hope to continue to make K4, K5 to K100 better in the next ten or twenty years."

The rules of the game have indeed changed. But the game is not over yet. This is a long journey. Kimi is on the way, and the entire industry is also on the way.