Moonshot AI’s latest verified funding round values the Kimi maker at more than $20 billion, but the bigger story is the speed of the repricing. Investors are treating Chinese open-weight models as infrastructure assets, not experimental chatbots.
Moonshot AI has become one of the clearest signals that China’s AI market is no longer waiting for perfect conditions. The Beijing company behind the Kimi large language model raised about $2 billion in May at a valuation above $20 billion, led by Meituan’s venture arm Long-Z Investments and joined by investors including China Mobile, according to Bloomberg’s report on the financing.
That number matters because of where Moonshot stood only months earlier. The company was valued at about $4.3 billion near the end of 2025, then moved through a new round early this year that pushed its valuation above $10 billion. By May, its financial adviser HF Capital said Moonshot had raised $3.9 billion over six months. That is not normal startup financing rhythm. It is a capital race.
The funding is also arriving in a market where Chinese AI companies have had to build under pressure. US chip restrictions have made access to the most advanced Nvidia hardware more complicated, but they have not stopped model development or investor appetite. If anything, they have sharpened the local argument: China needs strong domestic model companies, and the winners will need large amounts of capital for compute, talent and distribution.
Moonshot’s valuation is not being built only on national strategy. Kimi has become a serious product in its own right. The company is known for long-context models, and its newer Kimi releases have gained attention among developers looking for strong open-weight alternatives to Western systems from OpenAI, Anthropic and Google.
TechCrunch recently noted that Kimi K2.6 was the second-most used large language model on OpenRouter, a distribution platform that developers use to access different models. That is an important data point because usage is harder to dismiss than benchmark claims. Model demos can be polished. Leaderboards can move around. Developers spending tokens on a model suggests it is solving real work at a price and quality level they can justify.
The commercial side is beginning to show as well. South China Morning Post reported that Moonshot’s annual recurring revenue exceeded $200 million in April, citing HF Capital’s statement. That does not make Moonshot comparable to the largest US AI labs, but it gives investors something more concrete than future promise. The company sells subscriptions for Kimi and provides its model technology through APIs and enterprise channels.
That is why Meituan’s role in the round is worth watching. Meituan is not just another financial investor. It owns a massive consumer and merchant network across food delivery, local services and travel. If Kimi can be embedded into that kind of operating system, Moonshot gets something many model companies want but do not have: distribution tied to everyday transactions.
The valuation question is getting harder
The uncomfortable question is whether Moonshot’s private valuation is running ahead of the business. A $20 billion valuation against reported annual recurring revenue above $200 million is a very rich multiple, even in AI. Investors are not paying for today’s revenue alone. They are paying for the possibility that Kimi becomes a default layer for Chinese developers and enterprises.
That may still be rational. China’s AI field is crowded, but scale matters. DeepSeek has forced the industry to think harder about efficient training and open models. Alibaba has Qwen. ByteDance has Doubao. Zhipu AI and MiniMax have both drawn heavy investor attention and, according to recent market reports, have been valued above $30 billion in public or private-market comparisons. In that environment, Moonshot cannot afford to look undercapitalized.
The company’s backer list also explains why investors keep returning. Alibaba and Tencent were already among Moonshot’s supporters, along with names such as HongShan, ZhenFund, IDG Capital and 5Y Capital. When the largest Chinese technology platforms and top venture firms keep a model company funded, it tells the market that Kimi is viewed as strategically important, not merely fashionable.
Still, rapid step-ups create their own discipline. Every new round raises the bar for revenue, retention and enterprise adoption. It is one thing for developers to try Kimi because it is capable and cheaper than some Western alternatives. It is another thing for large companies to standardize on it for sensitive workflows, customer service, coding, research or internal automation.
Moonshot’s next phase will show whether China’s AI funding frenzy is rewarding durable products or chasing scarcity. The company has momentum, strong investors and a model family with real developer usage. What it now needs is proof that Kimi can turn attention into a repeatable business at the scale its valuation already assumes.
For the wider market, that is the point to watch. Chinese AI startups are no longer being valued as distant followers of Silicon Valley. They are being priced as possible platform companies in their own right. If Moonshot keeps converting Kimi usage into enterprise revenue, the May valuation may look like a staging post. If not, it will become an early test of how much patience investors really have for the next generation of AI labs.
Also read: Korea’s AI chip boom is pushing its bond market toward rate hikes • Nvidia and SK hynix are tightening their grip on AI memory • Forbes AI 50 shows AI startups are moving beyond model size
