Anthropic passes OpenAI, and the frontier reprices around who converts capability into revenue

The challenger became the front-runner, and it did it while the incumbent was still growing. That detail matters: Anthropic did not win by OpenAI stumbling. It won by growing faster from behind — and a coding tool did much of the lifting.

Anthropic is now valued at roughly $965 billion to OpenAI's $852 billion, with a revenue run-rate past $47 billion — far beyond earlier $10 billion projections. The reversal is recent, revenue-led, and it repriced a whole market.

Capability stopped being the scoreboard

The premium is repricing around who converts capability into durable enterprise revenue, not who tops a benchmark. Claude Code is the specific engine: coding assistance is the highest-retention enterprise workload of the cycle because it produces artifacts teams keep and integrate. The valuation followed the retention, not the leaderboard.

The timing fight makes the repricing concrete. OpenAI is weighing an IPO slip to 2027 after a rocky SpaceX debut, while Anthropic tracks a late-2026 listing and signals an expected first operating profit. A fall prospectus would put OpenAI's 2025 losses on the same page as a rival's profit — a side-by-side that is itself the repricing.

The public market will decide the math

Whichever lab lists first sets the comparable the others are measured against, which makes the IPO order a fight over who defines the frontier's valuation math. For the first time these companies will be priced directly rather than through venture rounds — and the market, not the benchmark, will render the verdict.

The frontier used to be won on capability. It is now won on conversion. Anthropic figured that out first.

CNBC — Anthropic tops OpenAI as most valuable AI startup → · Benzinga — OpenAI $1 trillion IPO ambition faces new timing test →