OpenAI's revenue run rate nears $70B as enterprise sales double
Sources familiar with OpenAI's finances told Axios the annualized pace rose more than 70% since July, while the company has not disclosed the costs behind it.
By Ryan Merket · Published
Primary source: Axios
Why it matters
OpenAI's reported pace puts it ahead of Anthropic on a headline revenue measure, but the figures are annualized estimates without a current expense or revenue-mix breakdown. The coming IPO filings will make the cost of that growth harder to avoid.
OpenAI, led by Sam Altman (@sama), is nearing a $70 billion annualized revenue run rate, with business-to-business revenue more than doubling since July, according to sources familiar with the company's finances who spoke to Axios. The figures mark a rapid reversal in the enterprise race with Anthropic, which had recently reported a higher run rate.
Altman co-founded OpenAI after co-founding location-based social-network company Loopt and later leading Y Combinator. His bet with OpenAI has been to build a broad AI business around consumer products, developer tools and corporate deployments. The newly reported growth suggests its consumer reach is increasingly translating into workplace sales, although the figures do not show how much of the increase comes from recurring contracts, API use or other products.
Axios reported that OpenAI's annualized revenue run rate increased more than 70% from the start of the third quarter, reaching almost $70 billion. Business-to-business revenue grew more than 100% over the same period. The report also says OpenAI added more revenue from consumers during the third quarter than it added during all of 2025. Those are company-financial figures relayed by sources familiar with them, not audited results or a published breakdown of revenue by product.
The distinction between revenue and a run rate is material. A run rate annualizes a recent pace of sales; it is not the same as revenue actually collected over a full year. The Axios report uses both "annual recurring revenue" and "annualized revenue run rate," but does not provide a methodology or specify how much of the figure is contractual recurring revenue. The reported near-$70 billion should therefore be read as a current pace, not as a forecast or a completed year's sales.
The pace has moved sharply in a short span. In August, Bloomberg reported that OpenAI was generating more than $40 billion in annualized revenue based on its then-current performance, citing people familiar with the matter. Axios's August report on Anthropic, citing figures reported by Bloomberg, put Anthropic above $65 billion by late July. The new OpenAI figure would put it ahead on that measure, if the estimates use comparable definitions and dates. Neither comparison establishes which business is more profitable or how much revenue is tied to durable customer commitments.
OpenAI had already said in its March 31st funding announcement that enterprise made up more than 40% of its revenue and was on track to reach parity with consumer revenue by the end of 2026. That announcement accompanied $122 billion in committed financing at an $852 billion post-money valuation. The September figures, if accurate, show corporate sales advancing quickly against that stated target, but OpenAI has not released a current revenue mix to confirm the shift.
The contrast with Anthropic is becoming clearer as both companies face public-market scrutiny. OpenAI said in June it had confidentially filed for a U.S. IPO, according to The Associated Press. Anthropic's prospectus, reviewed by Reuters, reported nearly $4.6 billion in 2025 revenue and more than $8 billion in operating losses, excluding certain financing-related accounting charges. It also disclosed $518 billion in future cloud, computing and infrastructure obligations. Those figures put the revenue race beside the costs required to keep it running.
OpenAI has not supplied expense details alongside the new run-rate figure, Axios reported. That leaves the central financial question unresolved: how much of the sales growth remains after the cost of compute, cloud capacity and model development. OpenAI said in March that its enterprise business was expanding and described compute as a strategic advantage; public investors will eventually need reported margins and cash flow to judge whether that advantage converts growth into a sustainable business.