It's hard to imagine that Anthropic's revenue curve would have turned out this way just a year ago.
Just this morning, according to Bloomberg, Anthropic disclosed to investors that by the end of July, the company's annual operating revenue (ARR) had reached $65 billion.
At the same time, Anthropic's initial revenue for the most recent quarter ending Q2 also surpassed $11.5 billion. This figure was only $787 million for the same period last year, which means that the latest quarterly revenue has reached about 14.6 times that of the same period last year.
For an AI company that was widely regarded as an OpenAI challenger a few years ago, this may be the fastest acceleration in its commercialization process to date.
For two months,
From $47 billion to $65 billion
It's worth pointing out that $65 billion doesn't represent that much revenue that Anthropic has actually made in a year.
The indicator used here is the "annual revenue operating rate". Simply put, it is based on the company's current revenue level over a period of time and estimates how much revenue it can generate for the entire year if this performance lasts for a year. This indicator is particularly suitable for looking at companies that are growing rapidly, as it can reflect current business changes more quickly than the revenue at the end of the fiscal year.
And the growth rate of Anthropic is now very evident.
At the end of 2025, its annual operating revenue was just over $9 billion; by May this year, the figure had surpassed $47 billion; and by the end of July, it had reached another $65 billion.
In other words, in just two months, Anthropic's annual operating revenue grew by about 38 percent.
On a longer-term scale, the change is more pronounced. According to Bloomberg, Anthropic will generate about $10 billion in full-year revenues in 2025.
Income growth has also begun to be reflected in profits.
Anthropic's latest quarterly initial revenue exceeded $11.5 billion, up significantly from $787 million in the same period in 2025. At the same time, the company has recorded an adjusted operating profit for the quarter. This is particularly noteworthy for front-end model companies that are still in the phase of large-scale computational input.
Over the past few years, the growth story of large model companies has often been accompanied by another aspect: the cost of training, the cost of reasoning, the simultaneous expansion of data center investment and the cost of talent.
Anthropic has begun to show positive adjusted operating profits, which at least indicates that its scale expansion is gradually showing operational leverage.
Claude Code is changing the pace of commercialization of Anthropic
Behind the Anthropic revenue surge, an inevitable variable is the enterprise market, particularly the programming and complex knowledge workplace.
Claude was initially seen more as a direct competitor to ChatGPT. However, over the past year, an apparent change in commercialization by Anthropic has been that it has begun to get deeper and deeper into the real workflow within the company.
Developers can directly have Claude read the code library, modify the file, execute the commands, check out the problems, and continue to complete the long chain software engineering tasks. As the use of Coding Agent in the enterprise expanded, model calls also evolved from an occasional conversation in the past into a continuous operation, consuming a large amount of Token workflow.
For model companies, the revenue generated by the two models is completely different.
A query may take only a few seconds to call a model; a software engineering agent that lasts for tens of minutes or even hours may repeatedly read the context, call tools, modify code, and re-verify results. This also explains why an enterprise-level agent is increasingly becoming the revenue engine for leading model companies.
According to data currently disclosed, Anthropic is already one of the most visible beneficiaries of this trend.
Bloomberg previously reported that OpenAI's recent annual operating revenue has surpassed $40 billion. By comparison alone, the latest disclosure by Anthropic has surged to $65 billion.
However, the two figures still need to be carefully compared directly. The statistical calibre of the annual revenue operating rate of different companies may not be entirely consistent, and differences in revenue confirmation methods and how enterprise contracts are calculated may result.
The IPO is getting closer and closer.
$65 billion is the most direct chip.
The timing of this figure is equally subtle. Anthropic is on the eve of its launch.
As previously reported by the Financial Times, investors at Anthropic expect the company to enter the open market in October of this year, with an IPO valuation of up to.Two trillion dollars or more.。
If eventually marketed at this valuation, Anthropic would surpass SpaceX, which went public earlier this year, to become the highest-valued IPO ever.
For Anthropic, which is on its way to capital markets, the $65 billion annual operating rate is obviously a very important sign.
AI companies have been highly valued in recent years, but the capital market will eventually return to a number of very traditional questions: how much revenue, how long growth can be sustained, when profits emerge, and whether massive computational investments can be converted into continuous cash flows.
Anthropic has long been regarded as the successor to OpenAI. It has now proven to potential investors that it has changed.
Modeling capabilities are still important, and enterprise market penetration, revenue growth and profitability are gaining increasing weight.
The environment currently facing Anthropic is not a smooth one either. In June this year, the company temporarily shut down access to the two advanced models Claude Fable 5 and Mythos 5 due to government regulatory requirements, only to recover after about two weeks of negotiations. Despite this, the company's latest revenue figures continue to grow rapidly, making this revenue curve more noteworthy.
The link is as follows: