How Much Revenue Does Anthropic Make Per Megawatt?
Dylan Patel says Anthropic's revenue has gone as high as $50 million per megawatt, while the base cost of compute runs around $10, $13, or $15 million per megawatt. Ten dollars spent on inference capacity can generate 50 dollars of revenue, and the profit can go into more training.
Last year, most of GDP growth in America was AI infrastructure. About a third of the compute coming online is for OpenAI and Anthropic, even when others build it and rent it to them. Capital spending is a little over a trillion dollars, and by 28 it is more than $2 trillion. The labs are taking a rising share, moving from tens of billions of dollars a year in spending to hundreds of billions, with forecasts of trillions of dollars a year toward the end of the decade on contracts they have begun signing.
Anthropic started turning a profit in Q2. OpenAI, it is believed, could potentially start turning a profit at some point in Q3, even with the rise of Codex and 5.6. A year ago the money was venture-funded losses. Over the last year and a half, margins have skyrocketed. GPT-4 served on NVIDIA Hopper GPUs generated negative gross margin for OpenAI. GPT 5.6, and Anthropic's Opus 5, Mythos, and Fable 5, now bring revenue well beyond an incremental $10 or $15 million per megawatt. More of the business is funded from its own revenue, while new capital still comes in to accelerate growth.
Dylan Patel – Anthropic & OpenAI will have most of the world’s compute by 2028 ·
From the transcript
But let's start with just where we are today. So walk me through lab compute and lab revenue right now and maybe projecting out a year or two. Yeah, so when we go back to last year, even at the end of the year, most of GDP growth in America was just AI infrastructure. And as we look towards this year, about a third of the compute coming online is for the labs, for OpenAI and Anthropic.
Now, it may be built by others and then rented to them, but at the end customer, it's them. As we go forward into the future, the numbers for computer ballooning, right, we're at a little bit over a trillion dollars of CapEx this year. As we go out into 28, it's gonna be more than $2 trillion.
The labs are also taking an increasing percentage of this. And so ultimately, you've got a very interesting situation where the labs are going from companies that spend tens of billions of dollars a year to hundreds of billions of dollars a year to forecasting to spend trillions of dollars a year even towards the end of the decade. And this is at least some of the contracts they've begun signing with their partners.