OpenAI Revenue Hits $10 Billion in Q3 2026 as Enterprise Demand Accelerates
Key Takeaways
- •OpenAI’s quarterly revenue rose approximately 50% to $10 billion in Q3 2026.
- •API demand, particularly for GPT-5.6 and Astra, was the company’s largest revenue driver, with advertising also contributing.
- •Enterprise customers now account for more than half of OpenAI’s revenue.
- •OpenAI’s annualized revenue run rate was about $50 billion at the end of September, compared with roughly $20 billion at the start of 2026.
- •Anthropic reported nearly $17 billion in Q3 revenue, exceeding OpenAI’s result by about $7 billion.

OpenAI generated $10 billion in revenue during the third quarter of 2026, reaching the milestone faster than many industry observers had anticipated. According to a report published by The Information on November 10, 2026, the figure represents growth of approximately 50% over the previous quarter. The result nonetheless trailed rival Anthropic, which announced nearly $17 billion in revenue for the same period — a gap of roughly $7 billion that offers a direct, side-by-side read on the two companies' commercial scale at the same point in time.
Where the Revenue Came From
The biggest driver was demand for OpenAI's application programming interfaces, or APIs — the plumbing that lets other companies wire OpenAI's models directly into their own products. Demand was particularly strong for the company's newer models, including GPT-5.6 and Astra. Advertising sales also gave the quarter a lift.
The customer mix is shifting as well. More than half of OpenAI's revenue now comes from enterprise clients, according to co-founder Greg Brockman — a milestone that puts corporate customers, rather than any other single segment, at the center of the company's growth story. Brockman also said OpenAI's annualized revenue grew 70% quarter-over-quarter in Q3, a different measure from the roughly 50% jump in actual quarterly revenue.
The Run-Rate Math
Run-rate figures extrapolate a single period's performance into an annualized number, which is why OpenAI can report $10 billion in quarterly revenue while citing a far larger annualized pace. By that measure, OpenAI was running at around $50 billion as of the end of September 2026 — more than double the approximately $20 billion recorded at the start of the year.
The run rate is expected to push past $70 billion by year's end, with enterprise sales expected to carry much of that load. Even so, the roughly $50 billion run-rate projection at the end of September reportedly caused some investor unease when it fell short of certain expectations — a reaction that illustrates how elevated the bar has become, given that the projection still represented more than double the pace recorded at the start of 2026.
Microsoft's OpenAI-Linked Revenue
The boom is also visible on Microsoft's books. The software giant said it generated $10 billion from its commercial engagements tied to OpenAI during its fiscal quarter ending September 30, 2026. That represents almost 11% of Microsoft's total quarterly revenue of $92.3 billion.
Across fiscal 2026, Microsoft's revenue from OpenAI-linked activities reached $24.1 billion — a full-year figure showing the contribution spans multiple reporting periods rather than a single quarter.
Anthropic Reports a Larger Quarter
Rival Anthropic announced nearly $17 billion in revenue for the same period, putting it ahead of OpenAI on quarterly revenue in Q3.
What This Means
Microsoft is the most direct way for investors in public technology companies to see OpenAI's momentum reflected in a reported earnings line. The almost 11% contribution to Microsoft's quarterly revenue makes the partnership a key variable in how the market evaluates the stock.
The figure to watch is the year-end run rate. If OpenAI clears the expected $70 billion mark, it would validate the enterprise-driven growth story Brockman is telling. Anthropic's quarterly disclosures, meanwhile, serve as the running benchmark for how that competitive gap evolves through the rest of the year.
This article is based on reporting by CryptoBriefing and was originally edited by Diego Almada Lopez. Source: CryptoBriefing