ByteDance, Alibaba and Tencent have ordered more than 3 million of Nvidia’s new RTX PRO 5500 graphics cards with a collective sticker price of over $35 billion, Reuters reported on 20 October 2026, citing two people familiar with the matter. Reuters described it as a rush for a less powerful but export-compliant chip amid US restrictions on trade with China.
Nvidia has told customers and partners it cannot guarantee that it can fill the orders because demand from Chinese technology companies far exceeds planned production, the two sources and a third person familiar with the matter told Reuters.
Business Pill · INFERENCE
A one-minute explainer of inference: running a trained AI model to produce answers. It teaches the general idea only and says nothing about any company in this story.
The key insight: As we read it, Reuters describes a rush for an export-compliant chip amid US restrictions: orders for more than 3 million RTX PRO 5500 workstation cards, among the most powerful Nvidia chips still available to Chinese buyers, with Nvidia telling partners it cannot guarantee it can fill them.
The three companies’ orders are for cards priced at about $11,400 each, two of the sources said, giving them a potential combined value of more than $35 billion, according to Reuters. Total orders from customers in China have exceeded 6 million, two of the sources said.
One source said Nvidia told partners they might receive only 5% to 10% of the quantities requested. Another said the company had indicated it could produce no more than 3 million RTX PRO 5500 cards globally in 2027, Reuters reported.
Nvidia has told customers to expect early samples in December, with commercial deliveries beginning in the first quarter of 2027, the sources said. Reuters said the orders had not been previously reported.
ByteDance, Alibaba and Tencent have ordered more than 3 million Nvidia RTX PRO 5500 cards with a collective sticker price of over $35 billion, Reuters reported on 20 October 2026, citing two people familiar. Order, output and allocation figures are as the sources told Reuters; the 34% tile is IDC data for data-centre AI chips (Reuters, 13 October 2026), a different category.
What the RTX PRO 5500 Is
The RTX PRO 5500 is a workstation chip, designed for powerful professional computers rather than the large data-centre clusters that use Nvidia’s most advanced AI processors, Reuters wrote. It is among the most powerful Nvidia chips still available to Chinese buyers.
Nvidia’s Blackwell-based B300 data-centre chip is not allowed to be sold to Chinese buyers under US export controls, while exports of its Hopper-based H200 are reviewed on a case-by-case basis and are subject to licensing conditions, according to Reuters.
Chinese companies are seeking to use the cards primarily for AI inference, the three sources said. The RTX PRO 5500 is similar to the RTX 6000D in performance, according to product specifications seen by Reuters. The RTX 6000D has since been discontinued and its market price in China has doubled, one of the sources said.
Why Supply Is Short
Nvidia’s inability to promise full deliveries reflects both a surge in demand and its manufacturing priorities, two of the sources told Reuters. Starting in 2027, Nvidia is expected to devote much of its capacity at Taiwan Semiconductor Manufacturing Co to Rubin, its next-generation platform for large-scale data centres, the two sources said. TSMC declined to comment.
An Nvidia spokesperson said Chinese producers of workstation chips and cards have experienced “unprecedented growth since 2022,” according to Reuters. “US firms continue to be restricted by a combination of outdated US export controls, which cover gaming products released nearly a half a decade ago, and China’s own limit on US imports,” the spokesperson said.
ByteDance, Alibaba and Tencent did not respond to requests for comment, Reuters said, nor did the US Commerce Department’s Bureau of Industry and Security or China’s Ministry of Industry and Information Technology.
Connecting the Coverage
On 13 October 2026, Reuters reported IDC data showing Nvidia’s share of China’s market for data-centre AI chips fell to 34% in the first half of 2026, which we covered in Nvidia’s China AI chip share falls to 34% as Huawei hits 28%. That figure is for data-centre AI chips, a different category from the RTX PRO 5500 workstation cards in today’s report.
What is new today is the scale of the orders, as Reuters reports it, and Nvidia’s warning that it cannot guarantee it can fill them.
The Structural Read
As we read it, the use case is running models, not training them: the three sources told Reuters the cards are mainly for AI inference, which generally requires less computing power than training.
As we read it, supply is the constraint the sources describe: total orders from China exceeded 6 million, while one source said Nvidia indicated it could make no more than 3 million cards globally in 2027.
As we read it, priorities matter too: two sources said Nvidia is expected to devote much of its TSMC capacity to Rubin from 2027, rather than to products like the RTX PRO 5500.
Nvidia spokesperson, as quoted by Reuters, 20 October 2026
“US firms continue to be restricted by a combination of outdated US export controls, which cover gaming products released nearly a half a decade ago, and Chinaโs own limit on US imports”
Three Implications
NVIDIA Reuters reports more than 3 million cards ordered by three firms at about $11,400 each; the company told partners they might receive only 5% to 10% of what they requested, one source said.
CHINESE AI BUYERS The sources told Reuters the cards are mainly for inference; the RTX 6000D, a similar card, has been discontinued and its price in China has doubled, one source said.
THE SUPPLY CHAIN Samples are expected in December and commercial deliveries from the first quarter of 2027, the sources said; TSMC declined to comment.
The framework puts it this way: “Training builds the brain. Inference is the brain working.”
As we read it, the Reuters sources describe inference demand: the three firms want the RTX PRO 5500 mainly for running trained models. The framework was published in March 2026 and does not discuss Nvidia’s China orders.
What Is Not Established
The order volumes, prices and production limits come from people familiar with the matter, not from Nvidia or the buyers. The Nvidia spokesperson’s statement quoted by Reuters does not address the order figures, and ByteDance, Alibaba and Tencent did not comment. Reuters does not report how many cards have shipped.
Business Engineer Framework
The State of the Inference Economy
A Business Engineer framework on why inference, not training, now carries most of AI compute and revenue.
ByteDance, Alibaba and Tencent have ordered more than 3 million RTX PRO 5500 cards worth over $35 billion, Reuters reported on 20 October 2026, citing people familiar, and Nvidia has told customers and partners it cannot guarantee it can fill the orders.
A note on sourcing. This piece is built on Reuters’ 20 October 2026 exclusive, read in full, which cites two and three people familiar with the matter. The 34% share is IDC data reported by Reuters on 13 October 2026. Nvidia, ByteDance, Alibaba and Tencent have not confirmed the order figures. Nothing here is a forecast of ours, and nothing here is financial or investment advice.
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.
Scroll to Top
Discover more from FourWeekMBA
Subscribe now to keep reading and get access to the full archive.
