Mech-Mind (09615): Beneath the Embodied Narrative, a Realistic Interrogation of a 30x PS Valuation

Title context: Mech-Mind (09615): Beneath the Embodied Narrative, a Realistic Interrogation of a 30x PS Valuation Text: Capital markets forever favor the next grand story. On September 1, 2026, Mech-Mind (09615) listed on the Hong Kong Stock Exchange as the "first listed embodied intelligence eyes-brain-hands company," with a 3,835x oversubscription that set a new Hong Kong IPO record for Siasun Robot&Automation. However, the contrast between the 3,835x pre-listing oversubscription frenzy and the post-listing share price pressure tears away a layer of the capital narrative. As of the September 28 close, its share price stood at HK$79.5, a discount of over 21% from its offer price (HK$101.7). On one side is the capital frenzy of a 3,835x oversubscription; on the other is the market cold shoulder of a sustained post-listing decline. Behind this valuation rift lies a question that must be answered: Was Mech-Mind's roughly 30x PS at issuance a reasonable pricing of the future of embodied intelligence, or an overly optimistic overvaluation? On this question, we may perhaps glean some clues from its latest financial report performance, its ability to deliver on technology, and its market prospects. The Two Sides of the Financial Report: Stellar Growth, but Profitability and Scale Effects Yet to Materialize On September 24, Mech-Mind delivered its first interim scorecard after listing: revenue, orders, and overseas income all surged across the board, with gross margin continuing to climbthe growth curve is dazzling enough. Yet parting the data's splendor, the other side is equally clear: profitability remains elusive, and scale effects are still a check that has yet to be cashed. In the first half, the company's revenue was RMB 237 million, up 54.7% year-on-year; newly signed orders were RMB 335 million, up 75.3% year-on-year; overseas revenue was approximately RMB 100 million, up 68.3% year-on-year, with overseas markets having already become an important growth engine. During the period, gross margin rose from 61.4% to 65.0%, with the product's own profitability continuing to optimize. Customer quality is equally outstanding, with a repurchase rate of 93.9% in the first half, and customers including more than 100 Fortune Global 500 companies, with business covering nearly 50 countries and regions. Beneath the dazzling growth curve, profit pressure remains heavy. In the first half, the company's operating loss widened to RMB 77.454 million, with R&D investment of RMB 80.1 million, up 72.9% year-on-year, and R&D expenses accounting for 33.8% of total revenue. Over a longer time span, from 2023 to 2025 the company's net losses were RMB 401 million, RMB 283 million, and RMB 360 million respectively, with a three-year adjusted cumulative loss of RMB 657 million; in Q1 2026 it continued to lose RMB 57 million. Although the adjusted loss narrowed slightly by 5.9%, revenue growth was 54.7%, while the operating loss simultaneously widened by 18.8%. It is worth noting that this set of data creates considerable tension with management's statement that "operating leverage is clearly emerging." The core logic of scale effects is that revenue expansion drives down unit fixed costs, and losses narrow at an accelerating pace as revenue grows. Mech-Mind's current performance shows that the scale dividend is far from truly arriving. Of course, changes in Mech-Mind's inventory also deserve attention: it rose from RMB 63.2 million at the end of 2025 to RMB 89.3 million at the end of the first half of 2026, an increase of 41.4%. Although lower than revenue growth, considering the company's high degree of product standardization, inventory buildup may still signal that downstream demand is not as optimistic as the order data suggests. Selling expenses are another signal. GMTEight observed that from 2023 to 2025, the company's selling expenses were RMB 186 million, RMB 162 million, and RMB 168 million respectively, continuously exceeding R&D expenses over the same period (RMB 119 million, RMB 109 million, and RMB 113 million). In 2023, the selling expense ratio once reached 102.7%, meaning that all revenue that year was still insufficient to cover selling costs. This raises a key question: Is the current high growth natural demand driven by product value, or orders bought through sustained high selling investment? Once selling investment contracts, whether growth can be maintained remains unknown. From the above, it is not hard to see that when Mech-Mind's order growth cannot outpace the speed at which losses widen, and when the declaration that "operating leverage is emerging" clashes with the reality of an 18.8% widening of operating losses, the rift between the company's valuation and fundamentals will not automatically close. The Fog of Technology: A "First Embodied Intelligence Stock" with 99% of Revenue from 3D Cameras The core of Mech-Mind's external narrative is building an integrated "eyes-brain-hands" solution for embodied intelligence Siasun Robot&Automation, but the Siasun Robot&Automation guidance business that currently contributes all revenue is essentially a mature set of 3D cameras and supporting software, serving traditional industrial Siasun Robot&Automation to complete object recognition, grasping, and loading/unloading in structured scenarios. This product line has been iterated for more than a decade and belongs to the mature solutions of the industrial vision track, not an embodied intelligence system aimed at humanoid Siasun Robot&Automation in open environments. Breaking down Mech-Mind's revenue structure, the answer is even more intuitive. In 2025, the company's total revenue was RMB 389 million, of which intelligent Siasun Robot&Automation guidance products contributed approximately RMB 361 million, accounting for 93%. Intelligent inspection and measurement products contributed approximately RMB 23.07 million, accounting for 5.9%. The two businesses together contributed about 99% of revenue. In the first half of 2026, intelligent Siasun Robot&Automation guidance products contributed RMB 216 million (91.14% of revenue), and intelligent inspection and measurement products contributed RMB 21.2 million (8.9% of revenue), together also contributing more than 99%. The two products truly branded with the name "embodied intelligence"the Mech-GPT multimodal large model and the Mech-Hand dexterous handtogether accounted for less than 1.1% of revenue. According to the prospectus plan, these two products are expected to begin scaled launch only at the end of 2026, and only in the second half of 2027 will they form "revenue of actual commercial scale." In other words, a company listed as the "first embodied intelligence stock" derives 99% of its revenue from an industrial 3D camera that has been sold for ten years. This is probably not the early stage of embodied intelligence commercialization, but rather a point at which embodied intelligence commercialization has not yet begun. This can be glimpsed from the technical paradox of "breakthrough progress." Management claimed in the financial report that "embodied brain technology has achieved major breakthroughs" and "validated the correctness and effectiveness of the investment direction." Yet in the first half of 2026, R&D expenses were RMB 80.1 million, up 72.9% year-on-year, accounting for about 33.8% of revenue. A paradox thus emerges: If the technology had truly achieved a "major breakthrough," R&D investment should have entered a stage of diminishing marginal returns; if R&D expenses are still growing rapidly, it indicates that the technology is far from converging. With only RMB 80 million in R&D expenses in the first half to simultaneously support Mech-GPT, Mech-Hand, and iteration of existing 3D vision productswhether this sum is enough to achieve so-called "breakthrough progress" in the "embodied brain" is worth questioning. And the company's founder Shao Tianlan also admitted in an interview with LatePost that mature paths for home scenarios and open-ended service industries are "not yet visible in the short term," and that some key technologies are "not yet fully clear." It can be seen that although Mech-Mind's solid foundation in industrial vision deserves recognition, mature industrial products cannot be directly equated with still-exploratory embodied intelligence. The company derives 99% of revenue from mature industrial 3D cameras, while embodied intelligence products such as Mech-GPT and the dexterous hand have yet to contribute substantive revenue. This means that Mech-Mind's grand narrative of 30x PS still has no corresponding revenue curve to endorse it. "Global No. 1," Born in a Narrow RMB 1.8 Billion Track For investors, the halo of "global No. 1" is a highly attractive label in the secondary market. Therefore, Mech-Mind's "global No. 1" also adds a stroke of thick ink to its valuation premium. According to data from CIC cited in the prospectus, the AI+3D vision-guided general intelligent Siasun Robot&Automation component market in which Mech-Mind operates had a total global size of only RMB 1.8 billion in 2025. Within this, Mech-Mind ranked first globally with a 22.1% revenue share. But it should be noted that Mech-Mind's global No. 1 crown sits atop an extremely small niche track, and this is the most easily overlooked undertone of the entire valuation logic. Broadening the view to the broader "vision-guided Siasun Robot&Automation (VGR) system" market, the scale is much larger. Data from Meticulous Research shows that the global VGR system market was USD 3.24 billion in 2026 and is expected to reach USD 16.92 billion by 2036, with a CAGR of 18.1%. VGR systems include 2D vision and 3D vision, covering hardware, software, and services, with application scenarios including assembly, quality inspection, and pick-and-place. The "AI+3D vision-guided general intelligent Siasun Robot&Automation component" in which Mech-Mind operates is an extremely small subset of the VGR marketit focuses on the "3D vision guidance component" of "general intelligent Siasun Robot&Automation." In the vision-guided Siasun Robot&Automation market, Mech-Mind chose an entry point with the fastest growth but the smallest scale. A global pool of RMB 1.8 billion means that even if it achieved a 50% share, revenue would be only RMB 900 million. Its RMB 389 million in revenue is already close to one-fifth of this niche market. Moreover, in terms of market growth, even though the forecast given by CIC is highly attractive: from 2025 to 2030, the CAGR of this niche market is as high as 43.2%, and the market size will expand to RMB 10.6 billion by 2030, horizontally comparing with the broader industry, this forecast also carries a radical flavor. According to data from Interact Analysis, over the same period the global machine vision market CAGR is only 7.2%, and China's machine vision market CAGR is about 10.1%. In other words, the forecast growth rate of Mech-Mind's track is 6 times the global market and more than 4 times the domestic market. And this high-growth forecast is also tied to a strong precondition not controlled by Mech-Mind: large-scale commercialization of general-purpose humanoid Siasun Robot&Automation from 2025 to 2030, driving the penetration rate of 3D vision components from 5.1% to 10.6%. Once the industrialization progress of humanoid Siasun Robot&Automation slows, the growth expectations of the entire niche market will have to be substantially revised down, and the underlying logic supporting the high valuation will also loosen accordingly. A big fish in a small pond is, after all, only a big fish in a small pond. This means that as a leader in a narrow track, no matter how fast Mech-Mind grows, it cannot outrun the arrival of the ceilingRMB 389 million in revenue corresponds to about one-fifth share, and the incremental space left for the "global No. 1" is visibly cramped. And the high-growth narrative supporting the valuation imagination is tied to the assumption of the suspense that is humanoid Siasun Robot&Automation commercialization, with self-evident risks. Conclusion So, is Mech-Mind's 30x PS overvalued? Based on the offer price, Mech-Mind's market value is approximately HK$12.7 billion, corresponding to 2025 revenue of RMB 389 million, a price-to-sales ratio of more than 30x. This valuation jump process is worth noting: the company's post-investment valuation in the Pre-IPO round was RMB 6.367 billion, corresponding to about 16x PS on 2025 revenue; the listing issue market value was approximately RMB 11.6 billion, with valuation rising 82% in just two months. The prospectus did not provide sufficient explanation for the huge valuation gap between the primary and secondary markets. And if it is said that Mech-Mind's valuation level is a premium the market pays for its global leadership, high gross margin structure, and the imaginative space of embodied intelligence, then Mech-Mind's 30x PS is already far above peers. By contrast, for mature industrial vision companies, Cognex's PS is about 9.5x, Opt Machine Vision Tech Co., Ltd. 9.4x, and Keyence only 6 to 7x. A simple calculation: Mech-Mind's revenue CAGR from 2023 to 2025 was 46.6%. If valuation returns to the industrial vision industry norm of 9-10x PS, corresponding to 2025 revenue of RMB 389 million, the reasonable market value range is only RMB 3.5-3.9 billion. Compared with the current market value, there is a significant premium. Of course, forward estimates cannot be directly equated with current valuation, but this comparison is enough to show that 30x PS implies multiple optimistic assumptionsmaintaining high revenue growth over the long term, Mech-GPT and the dexterous hand commercializing on schedule to open a new growth curve, and ultimately achieving scaled profitability. As long as any one assumption falls short of expectations, there is room for valuation downgrade. The market always has a simple cognition that high growth rates should enjoy high valuations. But the Mech-Mind case reminds us that growth rates cannot be priced separately from track space. High-speed growth in a narrow track can easily hit the ceiling quickly. Breaking issue on the first day of listing and a current share price discount of more than 21% may be a signal of the market's rational return.