In the quiet of the market's opening bell, the valuation reveals its true intent. On August 19, 2025, Unitree Robotics debuted on Shanghai's STAR Market at an opening price of 1,100 yuan per share—a 629.44% surge from its issue price of 150.80 yuan. The market capitalization hit 444.9 billion yuan ($61 billion). Behind this arithmetic lies a deeper signal: the capital market has just priced a new narrative for embodied AI, and not everyone is reading the code correctly.
Unitree is not a blockchain company. But as a Layer2 Research Lead who has spent years dissecting the disconnect between hype and protocol mechanics, I see a familiar pattern. The same forces that drove DeFi valuations to unsustainable highs in 2020 are now shaping the robotics sector. The difference is that Unitree's IPO is not a token launch—it is a public equity event with real regulatory gatekeeping. Yet the market's reaction is equally speculative.
The context is essential. Unitree, a Hangzhou-based robotics firm, is best known for its quadruped robots (Go2, B2) and humanoid prototypes (H1, G1). It is part of the "Hangzhou Six Little Dragons"—a group of startups the Chinese government has positioned as future industrial champions. The IPO was a landmark event: the first pure-play robotics company to list on the STAR Market. The issuing price of 150.80 yuan was set through book-building, implying a relatively conservative valuation. But the secondary market immediately priced in a 6.3x premium, reflecting extreme optimism about the "humanoid robot first stock" narrative.
Let me trace the core mechanics. The most revealing data point is the floating profit of Shunwei Capital, the investment vehicle linked to Xiaomi's Lei Jun. Shunwei's Astrend IV entity held 16.106 million shares at IPO, and the floating profit was reported at 15.2 billion yuan. Simple arithmetic shows that Astrend IV's average cost per share was approximately 56.4 yuan—about 37% of the issue price. That means the early investor saw a 1,850% return on paper at the opening price. This is not a criticism; it is a statement about the structural asymmetry between primary and secondary market pricing. The 629% IPO-day surge is a signal that the book-building process failed to capture the market's appetite for robotic exposure. Institutional investors priced a conservative thesis, while retail and algorithmic traders priced a revolution.
But the deeper question is whether the 444.9 billion yuan valuation is sustainable. Based on industry benchmarks, a robotics company at that market cap would need annual revenues of 55–90 billion yuan to justify a 5–8x price-to-sales ratio. Public data suggests Unitree's 2024 revenue was likely under 2 billion yuan. Even assuming 100% annual growth for five years, the valuation is pricing in a future that may not materialize. The IPO is a promise, not just a listing.
Now, the contrarian angle. The euphoria masks several blind spots that the original news coverage conveniently omitted. First, the floating profit is paper wealth. Chinese regulations require early investors to lock up shares for 1–3 years. The 15.2 billion yuan number is a headline, not a liquidity event. Second, the article failed to disclose the issuing price-to-sales ratio, which likely exceeds 50x—a level that would have dampened retail enthusiasm. Third, the competitive landscape is shifting. Unitree's strength is in hardware integration and cost control, but its AI capabilities—especially in end-to-end learning and large model integration—lag behind players like Figure AI (backed by OpenAI) and Tesla's Optimus. We audit not to judge, but to understand. In this case, the audit reveals a company that has mastered the mechanics of motion but not yet the intelligence that drives it.
There is also a geopolitical dimension. Unitree's overseas expansion faces scrutiny from Western regulators over data security and potential military use of its quadruped platforms. The company's listing on the STAR Market, rather than Hong Kong or the US, signals a strategic choice to maximize domestic policy support while minimizing cross-border risk. This is a double-edged sword: it protects against hostile regulation but limits access to international capital and talent.
What does this mean for the broader robotics ecosystem? The Unitree IPO creates a valuation anchor that will lift all boats in the near term. Competitors like Zhiyuan Robotics, Fourier Intelligence, and Galaxy General will likely see their own funding rounds reprice upward. But this is a double-edged sword: inflated valuations can lead to capital misallocation and a repeat of the 2021 NFT craze, where the signal was drowned by noise. Solitude clarifies the signal amidst the noise. The signal here is that the market is betting on physical AI as the next trillion-dollar theme. The noise is the assumption that any robot company can scale to that level.
Finally, the takeaway. The Unitree IPO is a watershed moment, but not for the reasons the headlines celebrate. It is a test of whether the capital market can sustain a valuation that is 5–8 years ahead of the fundamentals. The next 12 months will be critical: the first quarterly earnings report, the actual order book for humanoid robots, and the company's ability to close the AI gap with Western competitors. Authenticity is not minted, it is verified. Unitree has proven it can build robots. Now it must prove it can build a business that earns the trust of the market, not just the excitement of the moment.
