#KimiConfidentiallyFilesForHKIPO

Kimi Confidentially Files for a Hong Kong IPO: What It Really Means, Who Wins, Who Loses, and What I Think

There is a major misunderstanding around the phrase “confidential IPO filing.” It does not mean Moonshot AI has already gone public, that its shares are trading, or that the Hong Kong listing has been approved. A confidential filing simply allows a company to submit documents privately before the full public process begins, giving regulators and the company time to assess the listing, valuation and investor demand. For Moonshot AI, the developer of Kimi, the important point is that the company is clearly moving toward a Hong Kong listing, although the official HKEX filing should be the confirmation investors wait for. Reports indicate Moonshot is targeting a Hong Kong application by the end of September 2026, with the actual debut more realistically expected around Q4 2026 or Q1 2027.

The numbers behind Moonshot’s rise are extraordinary. The company was valued at around $4.3 billion in December 2025, reached approximately $10 billion after a January 2026 funding round, around $20 billion in May, and then completed a $3.5 billion Series F in July at a $35 billion post-money valuation. That means its valuation has increased roughly 8x from December in less than a year. Even more aggressive reports suggest Moonshot has been discussing a final pre-IPO round at a valuation of up to $50 billion, which would represent another roughly 43% increase from July. If that valuation becomes reality, Moonshot would enter the public market with expectations that are already extremely high before ordinary investors even get access to the stock.

The biggest reason behind this explosive repricing is Kimi K3, launched in July. The model reportedly uses a massive 2.8-trillion-parameter architecture with a mixture-of-experts design and a one-million-token context window. Its strong benchmark performance and open-weight strategy immediately attracted global attention and revived memories of the DeepSeek shock. The market reaction showed why this matters beyond Moonshot itself: investors began questioning whether frontier-level AI performance can continue to justify enormous spending on GPUs, data centres and AI infrastructure if increasingly capable Chinese models can deliver competitive performance at lower cost. That makes Moonshot’s IPO more than another technology listing; it could become a public-market test of the economic value of Chinese frontier AI.

Revenue is growing quickly, but this is where the valuation debate becomes complicated. Moonshot’s annualised recurring revenue reportedly increased from around $100 million in March to approximately $200 million in April and roughly $300 million by June. That growth is impressive, but compared with a $35 billion valuation, the implied valuation is already more than 100 times annualised revenue. At a potential $50 billion valuation, the multiple could approach 167x based on a $300 million run rate.

Even if revenue reaches $500 million, investors would still be paying around 100x revenue at a $50 billion valuation. This means the market is not simply pricing current business performance; it is pricing years of rapid growth, global enterprise adoption and successful commercialisation.

The bull case is powerful. Moonshot has a recognised AI brand, strong Kimi adoption, frontier-level model ambitions, open-weight distribution and rapidly increasing commercial revenue. Its restructuring ahead of a potential listing, participation from Chinese strategic capital and selection of major investment banks also indicate that the company is preparing seriously for the public market. Even more important are reports that Moonshot is negotiating revenue-sharing arrangements with Microsoft Azure, AWS and Google Cloud for K3.

If these partnerships become meaningful enterprise channels, they could transform Moonshot from a consumer AI company into a much larger AI infrastructure and model business. A successful IPO could also establish the valuation benchmark for other Chinese AI companies preparing to enter public markets.

But the bear case cannot be ignored. A 100x-plus revenue valuation leaves almost no room for execution mistakes. Kimi faces intense competition from Doubao, Qwen, DeepSeek and Tencent’s Yuanbao, while the AI market is experiencing rapid price compression as cheaper models appear. Moonshot has also faced compute-capacity pressure during periods of strong K3 demand, highlighting the infrastructure constraints Chinese AI companies can face because of US semiconductor restrictions. Its consumer-market leadership is not guaranteed, and geopolitical tensions create another layer of risk because Chinese AI companies face restrictions on US capital while simultaneously trying to work with major American cloud providers.

Hong Kong’s market environment adds another risk. The Hang Seng and Hang Seng Tech indexes have not been in a straight-line bull market, while Alibaba and other companies are bringing substantial AI-related capital requirements and equity supply to investors.

That means Moonshot could be entering the market at a time when investors are enthusiastic about AI but increasingly selective about valuation. Nvidia’s enormous revenue growth also demonstrates the difference between AI excitement and stock performance: extraordinary business growth does not automatically mean unlimited valuation expansion.

My view is simple: Moonshot AI going public could become one of the most important Chinese technology IPO stories of this cycle, but the company and the valuation must be judged separately. The technology is real, the revenue growth is impressive, and Kimi K3 has clearly changed how investors view Moonshot.

However, $35 billion already demands exceptional execution, while $50 billion would demand something close to flawless execution. I would therefore watch three things before becoming aggressively bullish: first, the official HKEX filing; second, whether the reported $300 million ARR continues growing after the initial K3 excitement; and third, whether Microsoft, AWS and Google partnerships turn into sustainable enterprise revenue. For retail investors, Gate’s KIMI Pre-IPO products may provide exposure to the story through mirror-note structures, but these are not actual Moonshot shares and carry cancellation, liquidity and profit-sharing risks.

The biggest mistake would be buying simply because headlines are pushing a $50 billion valuation.