Blog / Latest in AI

Latest in AI

OpenAI's quarterly revenue reportedly tops $10B for the first time

Unicorn News Desk ·

Abstract stepped blocks rising past a thin threshold line

Key facts

Q3 2026 revenue (reported)
More than $10 billion
Q2 2026 revenue (reported)
$6.7 billion
Quarter-on-quarter growth
Almost 50%
Adjusted operating margin
-109% in Q3, from -150% in Q2
Operating loss (reported)
$10.9 billion, from $10.1 billion

OpenAI’s quarterly revenue crossed $10 billion for the first time in the third quarter of 2026, Bloomberg reported on November 11, citing anonymous sources and crediting strong demand for the company’s recent AI models. If accurate, that is an almost 50% jump from the $6.7 billion OpenAI reportedly booked in the second quarter.

OpenAI has not published the figures itself, so this is a reported number rather than an official one. Another outlet, citing investor documents, puts the same quarter’s operating loss at $10.9 billion, which shows that faster sales growth has not yet closed the gap between revenue and costs.

What the Bloomberg report says about OpenAI’s third quarter

The headline figure is the first quarter in which OpenAI’s revenue has passed $10 billion. TechCrunch describes the result as OpenAI “bouncing back from the doldrums it experienced earlier this year”. The main details reported so far:

  • Revenue: more than $10 billion in the third quarter, up almost 50% from $6.7 billion in the second.
  • Growth rate: a clear acceleration. Growth between the first and second quarters was previously put at about 18%.
  • Margin: the adjusted operating margin narrowed from -150% in the second quarter to -109% in the third. An operating margin is operating profit or loss as a share of revenue, so -109% means operating costs exceeded revenue by slightly more than the revenue itself.
  • Run rate: OpenAI has reached an annual revenue run rate of $50 billion, according to Bloomberg. A run rate takes recent revenue and projects it over a full year.

TechCrunch attributes the recent gains to a mix of tighter pricing, discounts, better models, and a stronger focus on enterprise products and AI agents, meaning software that carries out tasks for people instead of only answering questions.

How $10 billion compares with OpenAI’s earlier quarters

The reported third quarter caps a steep climb. Memeburn, citing Wall Street Journal figures reported by Crypto Briefing, lists OpenAI’s booked revenue at $13.07 billion for all of 2025, $5.7 billion in the first quarter of 2026 and $6.7 billion in the second. If the Bloomberg figure holds, OpenAI’s third quarter alone would be close to 80% of its revenue for the whole of 2025.

PeriodRevenueBasis
Full year 2025$13.07 billionWSJ figures, cited by Memeburn
Q1 2026$5.7 billionWSJ figures, cited by Memeburn
Q2 2026$6.7 billionReported
Q3 2026More than $10 billionBloomberg, anonymous sources
Adjusted operating margin, Q2 to Q3-150% to -109%Reported
Operating loss, Q2 to Q3$10.1 billion to $10.9 billionInvestor documents, per BW Marketing World

For longer context: in June 2025, AIM reported that OpenAI had crossed a $10 billion annual recurring revenue mark, citing a spokesperson’s confirmation to CNBC, and that the company had lost $5 billion on $3.7 billion of revenue in 2024. A year and a half later, $10 billion is reportedly what OpenAI takes in during a single quarter.

A narrower loss margin, but a bigger loss in dollars

The profit picture depends on how it is measured. TechCrunch says OpenAI’s profitability “appears to have improved,” pointing to the margin moving from -150% to -109%. BW Marketing World, citing investor documents, reports the same margin change but adds that operating losses widened to $10.9 billion in the third quarter from $10.1 billion in the second.

Both readings hold at once. Revenue grew faster than costs, which narrowed the margin, but costs still grew, so the absolute loss went up. BW Marketing World places this within growing investor scrutiny of whether AI companies can turn fast revenue growth and heavy infrastructure spending into sustainable profits, and says the pattern reflects the financial pressure on companies investing in model development, computing infrastructure and data centres.

The timing matters because rival Anthropic has begun a roadshow for its proposed public offering, which could value it at about $1.4 trillion, according to reports. BW Marketing World expects that listing to act as a key benchmark for the AI sector. Unicorn API covered the IPO terms in Anthropic prices IPO at $70-$78 a share, valuing it at up to $1.4T.

Run rate versus booked revenue, and why OpenAI and Anthropic don’t line up

OpenAI’s growth still trails Anthropic’s on the headline numbers. Anthropic recently reported third-quarter revenue of $17 billion and has taken the lead in selling AI models to businesses, according to TechCrunch. A day before the Bloomberg report, Anthropic said its annual revenue run rate had grown by $10 billion within a month to $81 billion as of the end of October.

The two companies do not count the same way. Anthropic includes sales made by its cloud partners, while OpenAI reports only its share of some sales made through its partners. Memeburn spells out the mechanics: Anthropic books the full value of Claude sales made through Amazon, Google and Microsoft’s clouds, while OpenAI books only its own share after the 20% that goes to Microsoft, according to The Next Web’s summary of reporting by Bloomberg and The Information.

Two abstract scales weighing the same stack under different counting methods
OpenAI and Anthropic count revenue sold through cloud partners differently. · Illustration

The counting gap has already produced confusing headlines about OpenAI. In late September, Axios reported a run rate of almost $70 billion and CNBC put it at $68 billion; the Financial Times then reported that OpenAI had told investors its run rate was approaching $50 billion at the end of September, and a person familiar with the matter told CNBC the higher number had counted gross revenue from partnerships. Memeburn concludes that the gap came from how revenue is counted, not from lost sales. The $50 billion run rate Bloomberg now cites matches the lower, net figure.

Run-rate growth and quarterly growth also measure different things. A person familiar with OpenAI’s investor presentation told CNBC the run rate grew 77% in the third quarter, with the enterprise run rate up 107%, while chief financial officer Sarah Friar put growth from the end of June to the end of September at 70%, according to The Next Web, as cited by Memeburn. Those figures compare the end of one quarter with the end of the next. The almost 50% figure compares total revenue for the whole third quarter with the whole second quarter.

Microsoft’s side of the ledger

Microsoft’s filings give a partial outside view of the same business. Microsoft recorded $10 billion in revenue from OpenAI in its most recent quarter, according to its 10-Q filing as summarized by TechCrunch, which AI Weekly reports was nearly 11% of Microsoft’s total quarterly revenue of $92.3 billion. That is a separate $10 billion from OpenAI’s own reported revenue: it covers Microsoft’s commercial arrangements with OpenAI, including revenue-sharing payments.

Two interlocking abstract flows exchanging small tokens
Microsoft and OpenAI are tied by revenue sharing, cloud spending and an equity stake. · Illustration

Several points from that filing coverage help explain how OpenAI’s numbers are built:

  • OpenAI reportedly shares 20% of its revenue with Microsoft, and under an April 2026 agreement that revenue share continues through 2030, subject to a cap.
  • OpenAI committed to buying an additional $250 billion in Azure cloud services under an October 2025 agreement.
  • Microsoft reported $7.2 billion in outstanding receivables from OpenAI at the end of September, up from $6 billion at the end of June. Receivables are amounts billed but not yet paid.
  • Microsoft’s ownership stake slipped from about 25% in June to 24% in September.

The 20% revenue share is the reason OpenAI’s net figures look smaller than a gross count would show. The large cloud commitment is part of the cost side that keeps operating losses high even as revenue climbs.

What it means for developers building on OpenAI’s API

For teams building with OpenAI’s models through its API, the report points to a company growing quickly but still spending well beyond what it earns. Several points from the sources are relevant.

Pricing is a lever OpenAI is actively using

TechCrunch lists “tighter pricing” and discounts among the drivers of recent growth. That suggests OpenAI is adjusting how it charges in both directions. Builders whose costs depend on per-token prices or plan terms should expect continued changes, though no source describes specific upcoming price moves.

Enterprise and agent products are the priority

OpenAI’s increased focus on enterprise products and task-completing agents is cited as part of the turnaround, and the enterprise run rate reportedly grew 107% in the third quarter. Developers can reasonably expect more of OpenAI’s product attention to go to business and agent tooling. TechCrunch also notes that OpenAI says it has over a billion users, but that most are likely free users or subscribers to its cheapest plans.

Competition remains wide

OpenAI is competing not only with Anthropic but with models from Meta, Amazon, Microsoft and Google, along with many cheaper but still highly capable models from open source labs, especially in China. For builders, that competition is a practical reason to keep code portable across providers. Losses of $10.9 billion a quarter leave open how long OpenAI’s current pricing and discounts will last, and the sources do not settle that question.

Limits of the evidence

The central figure comes from Bloomberg citing anonymous sources, and TechCrunch frames it conditionally: “If true”. BW Marketing World attributes its loss figures to investor documents and an industry report. Neither is an audited filing from OpenAI. Earlier reporting on OpenAI’s run rate showed how easily figures can diverge depending on whether partner revenue is counted gross or net, so readers should check which basis a number uses before comparing OpenAI with Anthropic or with OpenAI’s own past results.

What to watch next

  • The year-end target. OpenAI has a year-end 2026 run-rate target of $70 billion or more. Memeburn calculates that, read as a net figure, this implies about 40% growth in the fourth quarter, roughly half the third-quarter pace.
  • Losses in dollars, not just margins. Whether operating losses stop growing even as margins improve.
  • Anthropic’s IPO. The listing is expected to set a public benchmark for AI company valuations, and its disclosures will allow a closer comparison with OpenAI’s reported numbers.
  • Microsoft’s next filing. Revenue from OpenAI and receivables owed by OpenAI provide a recurring outside check on OpenAI’s business.
  • Official confirmation. Whether OpenAI confirms the $10 billion quarter or publishes its own figures.

Frequently asked questions

How much revenue did OpenAI make in the third quarter of 2026?

Bloomberg, citing anonymous sources, reports that OpenAI's quarterly revenue crossed $10 billion for the first time. OpenAI has not published the figure itself.

Is OpenAI profitable?

No, according to the reports. Its adjusted operating margin reportedly improved from -150% in the second quarter to -109% in the third, but operating losses grew in dollar terms to $10.9 billion from $10.1 billion.

How does OpenAI's revenue compare with Anthropic's?

Anthropic recently reported third-quarter revenue of $17 billion and says its annual run rate reached $81 billion at the end of October, against $50 billion for OpenAI per Bloomberg. The two count partner sales differently: Anthropic includes sales made by its cloud partners, while OpenAI reports only its share of some partner sales.

Why did OpenAI's revenue growth speed up?

TechCrunch credits tighter pricing, discounts, improving models and a greater focus on enterprise products and AI agents. Bloomberg's report attributes the jump to strong demand for OpenAI's recent models.