Suzhou Denglin Technology, a Chinese maker of AI graphics processors, raised HK$1,414.9 million in gross proceeds after pricing its Hong Kong IPO at HK$231.00, the bottom of its HK$231.00 to HK$288.00 range, according to its allotment results on HKEXnews. Net proceeds of HK$1,312.4 million are HK$167.6 million, or 11.3%, less than the HK$1,480.0 million its prospectus estimated at the mid-point, by our arithmetic.
The shares opened at HK$229.00 on Tuesday, 10 November, 0.9% below the offer price, and stood at HK$219.80 at 09:46 Hong Kong time (01:46 UTC), down 4.8%, according to delayed HKEX quotes. Jiangsu Xinde Semiconductor, a chip-packaging company that listed the same morning at HK$8.88, opened 20.5% higher and was up 30.7% at the same time. These are intraday prices; the session was still open.
Business Pill · Concentration Explained
An evergreen Business Engineer explainer on market concentration. It teaches the general idea only and does not discuss Denglin or Xinde.
The key insight: Denglin, a Chinese AI-GPU maker, priced at the HK$231.00 floor and netted HK$1,312.4 million, HK$167.6 million or 11.3% below its prospectus mid-point estimate by our arithmetic. At 09:46 Hong Kong time it was 4.8% below its offer price, while Xinde, a chip packager listed the same morning, was up 30.7%.
Denglin sold 6,125,000 H shares, 95% of them in the international offering, its announcement shows. The Hong Kong public offering was 2.11 times subscribed and the international offering 1.34 times, across 37 placees.
Five cornerstone investors took 2,798,025 shares, 45.68% of the offer, worth about HK$646.3 million at the final price by our arithmetic. Hua Yuan International took 1,516,450 shares and Sky Giant Investment 1,000,875; Suzhou Chinese Consortium, ZGD Capital and Everbright Wealth Management took the rest. Hua Yuan and Suzhou Chinese Consortium are listed as existing shareholders or their close associates, placed under an exchange consent.
Xinde sold 106,562,500 H shares for gross proceeds of HK$946.28 million and net proceeds of HK$881.62 million, its results show, after pricing inside its HK$8.60 to HK$9.60 range. Its Hong Kong public offer was 9.84 times subscribed, against 1.13 times for the international offer. The two listings raised HK$2,361.2 million gross together, by our arithmetic.
% change from each final offer price (HKEX allotment results): at the open and at 09:46 Hong Kong time (01:46 UTC) on 10 November, from delayed HKEX quotes. Intraday, not closing prices; percentages are our arithmetic.
What Denglin Makes
Denglin describes itself as a “general-purpose graphics processing unit (GPGPU) provider” for enterprise AI in China. Its prospectus says it was the third-largest company in China’s enterprise AI GPGPU market by 2025 revenue, with a 1.0% share, according to Frost & Sullivan, and that its cards are “fully compatible with CUDA”, Nvidia’s software platform.
It listed under the exchange’s Chapter 18C rules as a pre-commercial company, one that cannot meet the revenue requirement in Rule 18C.03(4), the prospectus says. Revenue was RMB171.3 million in 2025 and RMB200.4 million in the four months to 30 April 2026, more than the whole of 2025. Its loss narrowed to RMB16.5 million in those four months from RMB108.3 million a year earlier.
Where Nvidia Fits
The prospectus, citing Frost & Sullivan, says that in 2025 “NVIDIA and AMD collectively accounted for over 90% of the market” for public cloud AI GPGPUs in China, and that export restrictions limited the availability of certain of their products there. Denglin’s core market is enterprise AI; it says it is seeking to expand into the public cloud market.
The risk factors name U.S. Entity List and foreign direct product rules as a possible constraint on its access to technology and components. Nvidia’s own China exposure was the subject of our reading of its 10-Q at 8.2%, after its chief executive called China chip curbs a mistake.
The Structural Read
As we read it, the two books split along the same line as the first trades: Xinde’s retail tranche was 9.84 times covered and Denglin’s 2.11 times, and cornerstones took 45.68% of Denglin’s offer.
The framework argues that the physical supply chain under AI is “creating geopolitical leverage that operates independently of who has the best models”.
As we read it, Denglin’s own risk factors describe that leverage from the buyer’s side: Entity List and foreign direct product rules that could limit its access to technology and components. The framework predates the listing and does not discuss it.
What Is Not Established
These are intraday prices read from delayed HKEX quotes at 09:46 Hong Kong time, not closing prices, and they can move before the close. We read only the headline of Reuters’ report on the debuts, which called them mixed. Denglin’s figures are before any exercise of its over-allotment option, so its proceeds could still rise. Xinde’s prospectus says it has recorded losses, including RMB483.1 million in 2025.
This Week In The AI Supercycle
The Great Convergence of Capital, Power, and Geopolitics
This week’s issue: for two years the AI build-out was priced as a supply story, and last week as a credit story. This week it was priced as a permission story, and the answer arrived as a cost, not a ban.
Denglin priced its IPO at the floor and netted HK$1,312.4 million, HK$167.6 million below its prospectus mid-point estimate by our arithmetic, and traded 4.8% below its offer price at 09:46 Hong Kong time. Xinde, which packages chips, was up 30.7% at the same moment.
A note on sourcing. Offer prices, proceeds, subscription levels and cornerstone allocations are from the companies’ final offer price and allotment results announcements on HKEXnews, and business and financial details from their prospectuses. Share prices are delayed HKEX quotes as of 09:46 Hong Kong time (01:46 UTC) on 10 November 2026, read while the session was open, so they are intraday figures, not closes. We read only the headline of the Reuters report. The proceeds gap, the cornerstone value, the combined proceeds and the percentage changes are our arithmetic.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
