Five Taiwanese companies that build AI servers reported October 2026 revenue up 46% to 137% from a year earlier, according to their monthly filings on Taiwan’s Market Observation Post System (MOPS). Gigabyte rose 136.72% to NT$67.0 billion and Wistron 117.56% to NT$402.6 billion.
By our arithmetic, the five reported NT$2.02 trillion of October revenue combined, up 60.9% from NT$1.26 trillion a year earlier. Quanta and TSMC had not filed October figures when we checked at about 08:35 UTC on 7 November; the filing deadline is the 10th.
Business Pill · Concentration Explained
A short explainer on concentration. It teaches the general idea only and does not discuss these companies or their filings.
The key insight: By our arithmetic, five Taiwanese AI server makers filed NT$2.02 trillion of October revenue, up 60.9% from a year earlier, in a week when AI shares fell sharply.
Wistron reported October revenue of NT$402.61 billion, up 117.56% from NT$185.06 billion a year earlier, and NT$3.29 trillion for January to October, up 99.53%. Its filing gives the reason as “ไธปไฟๅฎขๆถ้ๆฑๅขๅ ๏ผ่ผๅปๅนดๅๆๆ้ท”, mainly higher customer demand (MOPS monthly revenue filing).
Wiwynn reported NT$165.13 billion, up 80.94%, and NT$1.15 trillion for the year to date, up 54.03%; its filing cites “ไธป่ฆไฟๅฎขๆถ้ๆฑๅผทๅ”, mainly strong customer demand. Gigabyte reported NT$67.04 billion, up 136.72%, and NT$471.0 billion year to date, up 70.72%, citing “AI่จๅฎ้ๆฑๆขๆบ”, strong demand for AI orders.
Hon Hai, also known as Foxconn, reported NT$1.31 trillion, up 45.73%, and NT$8.98 trillion year to date, up 40.40%. Its revenue also includes non-AI products such as consumer electronics. Inventec reported NT$79.97 billion, up 46.02%, and NT$822.2 billion year to date, up 43.08%.
October 2026 revenue change from a year earlier at five Taiwanese AI server makers, from their MOPS monthly filings: Gigabyte +136.7%, Wistron +117.6%, Wiwynn +80.9%, Inventec +46.0%, Hon Hai +45.7%. Hon Hai’s revenue includes non-AI products; Quanta and TSMC had not filed October.
Our Numbers
By our arithmetic, the five companies’ October revenue of NT$2,020.1 billion is 60.9% above their combined NT$1,255.1 billion a year earlier and 7.8% above their combined NT$1,873.8 billion in September, using September figures from the Taiwan Stock Exchange’s open data.
Without Hon Hai, the other four reported NT$714.7 billion, up 98.9% from a year earlier, by our arithmetic. Month on month, Hon Hai rose 12.66%, Gigabyte 6.24% and Wistron 5.73%, while Wiwynn fell 4.34% and Inventec 18.93%, by our arithmetic from the September filings.
Analyst Dan Nystedt, citing media reports, put Wistron’s October revenue at about US$12.3 billion and said it was the company’s second-highest monthly figure on record.
As we read the filings, Gigabyte and Wistron grew fastest, at 136.72% and 117.56%, and only Gigabyte’s filing names AI orders as the reason.
As we read the totals, Hon Hai is 64.6% of the five companies’ combined October revenue by our arithmetic, and its figure includes large non-AI businesses, so the four others are the cleaner read: NT$714.7 billion, up 98.9%.
As we read the month-on-month figures, the picture is mixed: Hon Hai, Gigabyte and Wistron rose from September while Wiwynn and Inventec fell, by our arithmetic.
Gigabyte, October 2026 monthly revenue filing on MOPS
“AI่จๅฎ้ๆฑๆขๆบ”
Three Implications
FOR READERS TRACKING AI DEMAND By our arithmetic, the five companies’ October revenue was 7.8% above September and 60.9% above October 2025.
FOR READING HON HAI Hon Hai’s NT$1.31 trillion includes consumer electronics and other non-AI products; the filing does not split them out.
FOR WHAT IS STILL MISSING Quanta and TSMC had not filed October revenue when we checked; the monthly filings are due by the 10th.
The framework sets out to map the financing of AI: “How the money flows, where it cracks, what it buys, who can absorb it, and what happens when revenue can’t catch up.”
The framework does not discuss these companies’ monthly revenue filings.
What Is Not Established
The monthly filings give total revenue only. They do not split AI servers from other products, name customers, or give margins. Hon Hai’s figure in particular includes large non-AI businesses. Quanta and TSMC had not filed October revenue when we checked, so they are not in the totals.
This Week In The AI Supercycle
Anthropic’s Reality Check, The Capital Squeeze, & The New World Order
This week’s issue: for two years the AI build-out was priced as a growthstory. This week it started being priced as a credit story.
Gigabyte, Wistron, Wiwynn, Inventec and Hon Hai filed October revenue up 46% to 137% from a year earlier. By our arithmetic, the five together reported NT$2.02 trillion, up 60.9%, and the four without Hon Hai NT$714.7 billion, up 98.9%.
A note on sourcing. October revenue figures and year-on-year changes come from each company’s monthly revenue filing on Taiwan’s Market Observation Post System, read on 7 November 2026; September figures come from the Taiwan Stock Exchange’s open data. Combined totals, month-on-month changes and the figures without Hon Hai are our arithmetic. Quanta and TSMC had not filed October revenue when we checked.
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.
