Anthropic is finalizing a $15 billion pre-IPO credit facility while its chief business officer commits to value-based pricing over OpenAI's price cuts ahead of its public listing.
Anthropic is finalizing a $15 billion pre-IPO credit facility while its chief business officer commits to value-based pricing over OpenAI's price cuts ahead of its public listing.

Anthropic PBC is finalizing a $15 billion revolving credit facility ahead of its IPO filing, a sixfold expansion from last year's $2.5 billion line that gives the Claude maker a liquidity backstop as it prepares to go public. The facility, led by Morgan Stanley with Goldman Sachs, JPMorgan Chase and Citigroup in prominent roles, exceeds the roughly $10 billion target Bloomberg reported in August.
"I have no interest in buying market share through price cuts," Paul Smith, Anthropic's chief business officer, said in an interview. "I'd rather focus on customers and how they extract maximum value from the models."
The syndicate extends well beyond the four lead banks. Barclays and Wells Fargo hold key roles, with Bank of America, Deutsche Bank, Royal Bank of Canada and UBS also high in the lineup. Bank of Montreal, BNP Paribas, Credit Agricole, Mizuho, Mitsubishi UFJ, Sumitomo Mitsui and Toronto-Dominion round out the roster. The four lead lenders are also guiding the IPO, Bloomberg News has reported.
The financing clears a hurdle before Anthropic's confidential IPO filing, which Reuters reported in June. The company is seeking to raise as much as or more than SpaceX, whose record-setting $86.2 billion debut in June included the over-allotment. US listings this year have raised $160.6 billion, the most since 2021, according to Bloomberg-compiled data.
Pricing divergence sharpens
Smith's stance marks a deliberate departure from OpenAI, which cut prices on multiple AI models this summer in response to competitive pressure from US and Chinese developers. Anthropic, widely regarded as the capability benchmark in frontier AI, instead released Fable 5.1 — its latest model focused on coding and scientific tasks — at pricing unchanged from the prior version.
The company did trim costs in one specific area: cache reads fell 75 percent, cutting typical workload costs by about 25 percent versus Fable 5, with complex coding and highly agentic tasks seeing reductions up to 45 percent. That targeted efficiency gain lowers effective costs for heavy users without triggering a broad price war.
The pricing philosophy reflects Smith's view that enterprise customers buy outcomes, not tokens. He joined Anthropic more than a year ago to lead enterprise sales and has since tripled the sales team. Annualized revenue has grown more than tenfold over the same period, with the company now on track to generate over $65 billion on an annualized basis — up more than sevenfold from its pace at the end of last year, Bloomberg reported.
Enterprise adoption matures
Smith describes enterprise adoption as a two-phase pattern. Initially, companies open access broadly, encouraging employees to experiment with Claude to overcome organizational inertia. Then they tighten controls, managing usage by role, function and budget allocation.
"In almost every case, they're still growing, still increasing spend on Anthropic, still using more Claude," Smith said. "But in this new phase, usage is more controlled. It's a natural maturation of the organizational learning process."
Data from Yipit and Ramp shows enterprise sales growth is still rising, though the pace has slowed from the peak of the Claude Code breakout period. Smith attributes this to the normal rhythm of enterprise adoption cycles rather than demand weakness.
Product-wise, Claude Code has spread rapidly through developer teams because software engineers are highly autonomous and can scale usage across an organization quickly. Claude Cowork, aimed at knowledge workers in finance, legal and research, has grown faster in early stages but requires more integration support and partner development.
"98 percent of enterprise users are not software engineers," Smith said. "There's a large amount of AI use cases in the enterprise we haven't unlocked yet."
The $15 billion revolver serves primarily as a liquidity backstop rather than a borrowing commitment. Having it in place can lower perceived near-term financing risk during IPO due diligence, potentially influencing how much new capital the company sells versus how much existing shareholders offload — a trade-off that affects dilution and float. For the banks, the economics cut both ways: they earn commitment fees even if the line goes unused, but they also take on contingent exposure that may need to be syndicated out, which helps explain the broad roster.
Smith declined to comment on the IPO timeline but said going public won't change how Anthropic operates. "It's not the finish line, it's just an event," he said. "Our mission and how we run the business won't change."
Anthropic raised $65 billion in May at a $965 billion post-money valuation and $30 billion in February at $380 billion. Prediction markets currently imply a market cap between $1.75 trillion and $2 trillion at listing, according to Vera data.
This article is for informational purposes only and does not constitute investment advice.