Anthropic closed out July with an annualized revenue run rate north of $65 billion, according to figures the company shared with investors, and the number says less about a single quarter than it does about a company that seems physically incapable of slowing down. Eighteen months ago, Anthropic was a research lab known mostly for building a very good chatbot that lost to ChatGPT on name recognition. Today, it’s burning through revenue milestones so quickly that analysts are struggling to keep their spreadsheets up to date.

The Number That’s Making Wall Street Pay Attention

Let’s put $65 billion in context, because the figure alone doesn’t quite capture how strange this trajectory has been. Anthropic’s run rate — a metric that takes recent revenue and projects it out over a full year — sat at roughly $9 billion at the end of 2025. By May of this year, it had jumped to $47 billion. Two months later, it’s $65 billion. That’s an $18 billion jump in eight weeks, which is more annualized revenue than most public software companies generate in their entire existence.

The underlying quarterly numbers back up the story. Anthropic’s preliminary second-quarter revenue came in above $11.5 billion, more than double what it pulled in during the first quarter and roughly fourteen times what it made in the same period a year ago. For a company that was essentially pre-revenue three years ago, that’s not a growth curve; it’s closer to a growth cliff — straight up.

Anthropic’s run-rate climb, quarter by quarter:

  • End of 2025: ~$9 billion
  • May 2026: $47 billion
  • End of July 2026: $65 billion+
  • Investor projection for year-end 2026: $100–120 billion

That last number is the one worth sitting with. If investors are right, Anthropic could nearly double its current run rate again before the calendar flips to 2027.

What’s Actually Driving This

The obvious question is where all this money is coming from, and the answer is less about consumer subscriptions and more about businesses quietly rewiring how they build software. Enterprise customers now account for the overwhelming majority of Anthropic’s revenue, and the company’s coding-focused product, Claude Code, has become the centerpiece of that shift. What started as a developer tool has turned into one of the fastest-scaling products in software history, and it’s dragging the rest of Anthropic’s enterprise business along with it.

The customer numbers tell the same story from a different angle. Anthropic now counts more than 300,000 business customers, and the number of accounts spending over $1 million a year has grown sharply in just the past several months. That’s the kind of expansion that shows up not just in headline revenue but in the durability of it — six and seven-figure enterprise contracts tend to stick around longer than a consumer subscription someone forgets to cancel.

It’s Not Just Coding Anymore

Anthropic has also been pushing its agentic tools beyond software engineering and into areas like finance, legal work, and general office tasks — the kind of “have an AI actually do the task, not just describe how to do it” pitch that’s become the industry’s favorite phrase this year. Whether that expansion holds up as well as Claude Code has is still an open question, but it’s clearly part of the bet behind these numbers.

Anthropic vs. OpenAI: The Gap Is Widening

For most of the last three years, this conversation defaulted to OpenAI. That’s changing, at least on the revenue side. OpenAI’s own annualized run rate recently reached roughly $40 billion, up from about $20 billion at the end of last year — genuinely strong growth by any normal standard, and yet Anthropic is now pulling ahead by a wide margin using the same yardstick.

It’s worth a caveat here: the two companies don’t necessarily calculate “annualized revenue” identically, and run-rate figures are projections, not audited full-year results. But even accounting for that, the gap has investors talking about Anthropic differently than they were a year ago. OpenAI still dominates in raw consumer traffic and brand recognition — ChatGPT’s user base dwarfs Claude’s by a wide margin — but Anthropic has built the more profitable, enterprise-heavy business, and profitability has become the metric everyone in this space is suddenly obsessed with.

The IPO Everyone’s Been Waiting For

None of this is happening in a vacuum. Anthropic filed confidential paperwork for an initial public offering back in June, and the company has reportedly been meeting with prospective investors ahead of a listing that could arrive as soon as September or October. Big-name banks are already lined up to run the offering, and the pitch to Wall Street basically writes itself at this point: a company that went from roughly a billion dollars a year in revenue to a projected $100 billion-plus run rate in under two years.

Anthropic’s valuation has climbed in step with the revenue. The company closed a $65 billion funding round back in late May at a $965 billion valuation, and some investors are now floating the possibility that a public listing could push that number above $2 trillion. Whether the market actually supports a figure that large once the company is trading is a separate question — plenty of hyped IPOs have landed with a thud after the roadshow ends — but the appetite going in is clearly real.

What This Means

A $65 billion run rate doesn’t mean Anthropic banked $65 billion this year — it means the company is currently operating at a pace that, if sustained, would produce that much revenue over twelve months. Run rates can be misleading in both directions: they can undersell a company that’s about to slow a hot streak, or they can overstate one that’s riding a temporary surge. Given how consistently Anthropic has beaten its own growth targets over the past two years, though, betting against this trajectory hasn’t been a smart move so far.

The bigger takeaway is what this says about enterprise AI spending broadly. Companies aren’t just experimenting with chatbots anymore — they’re routing real budget toward tools that write code, handle workflows, and increasingly act rather than just answer. Anthropic happened to build the product businesses wanted most at exactly the moment they were ready to pay serious money for it. Whether that holds through an IPO, a possible market correction, or the inevitable moment when growth this fast has to slow down is the story that’s just getting started.