OpenAI just had its best quarter ever, and it still isn’t winning the race that matters most. According to TechCrunch’s summary of a Bloomberg report, the company’s quarterly revenue crossed $10 billion for the first time. The numbers come from anonymous sources, so treat them as unconfirmed.
If accurate, that is a jump of almost 50% from the $6.7 billion reportedly booked in the second quarter. The prior quarter-over-quarter growth rate was about 18%, so this is a real acceleration. Profitability also moved in the right direction. The adjusted operating margin reportedly improved from -150% in Q2 to -109% in Q3. Still deeply negative, but narrowing.
Context matters here. OpenAI wasn’t exactly stalling before. Growing 18% on a base of roughly $5.7 billion a quarter is huge by software standards. The problem was Anthropic, which has taken the lead in enterprise AI and recently reported Q3 revenue of $17 billion. OpenAI says it has over a billion users, and that gave it a huge early lead in consumer mindshare. But most of those users are probably on free or cheapest-tier plans, which don’t pay the bills the way corporate contracts do.
So what changed? The report points to tighter pricing, discounts, better models, and a sharper push into enterprise products and AI agents that complete tasks on a user’s behalf. That last piece is the real strategic shift. Chat subscriptions are a crowded, price-sensitive business. Agents embedded in company workflows are stickier, and that’s where Anthropic’s Claude has been winning.
The run-rate figures show the gap. OpenAI is reportedly at a $50 billion annual run rate. Anthropic said its own run rate rose $10 billion in a month to $81 billion as of the end of October, and it is reportedly seeking a $1.4 trillion valuation in an IPO that could come soon. One caveat: the two companies count revenue differently. Anthropic includes sales made through its cloud partners, while OpenAI reports only its share of some partner sales. That makes the comparison less lopsided than it looks, though not dramatically so.
For developers and founders, the takeaway is that neither company has a safe position. Meta, Amazon, Microsoft, and Google all have competing models, and open source labs, especially in China, keep shipping cheaper models that are good enough for many jobs. That pressure helps explain OpenAI’s discounting. This quarter is a legitimate rebound and worth noting. But it’s a rebound in second place on enterprise revenue, and the next few quarters will show whether the agent strategy can close the gap or only slow it.
