Nebius Secures $775M Debt Facility to Expand AI Cloud Without New Share Issuance

Nebius has secured approximately $775 million through its first senior secured debt facility, using deployed GPU infrastructure and contracted customer cash flows as the foundation for the financing. The proceeds are intended to accelerate the global expansion of its full-stack AI cloud platform without issuing new shares as part of this transaction, according to the company’s July 17 financing announcement.

The deal matters because it gives Nebius a financing structure that can potentially be repeated across additional long-term customer deployments. At the same time, investors should separate the debt transaction from Nvidia’s 9.3% beneficial ownership disclosure: the latter includes existing shares and warrant-related shares, while the new facility is debt backed by infrastructure and contracted cash flows, not an equity raise.

What Nebius raised and how the facility is structured

The facility is approximately $775 million and matures on October 31, 2030. Nebius says it is priced at SOFR plus 2.50%, meaning its interest expense will move with the secured overnight financing rate rather than remain fully fixed for the life of the loan.

The collateral structure is central to the transaction. The financing is backed by deployed GPU infrastructure and contracted cash flows from an investment-grade customer. Nebius also says that the facility, together with cash flows under the customer agreement, covers more than 100% of the capital expenditure required to deploy the underlying GPU infrastructure.

This is different from borrowing against a general corporate balance sheet. The lender is evaluating a specific operating asset and the revenue associated with it, which can make the risk easier to underwrite when the customer contract, hardware deployment and servicing history are sufficiently clear.

Why this is non-dilutive, and what that does not mean

Debt financing is described as non-dilutive because Nebius is not selling new ordinary shares to fund this $775 million facility. Existing shareholders therefore avoid the immediate increase in share count that would normally accompany a conventional equity offering.

However, non-dilutive does not mean cost-free or risk-free. Interest payments, principal repayment and security over assets create fixed obligations. If the pledged infrastructure produces less cash flow than expected, or if customer deployment schedules change, the company may have less financial flexibility than it would after raising equity.

Investors should also avoid treating every Nebius capital transaction as equivalent. In March 2026, Nebius disclosed a separate $2 billion private placement with Nvidia involving a pre-funded warrant for 21,065,936 Class A ordinary shares, as documented in the company’s SEC Form 6-K filing. That warrant-related arrangement has a potential equity impact; the July secured debt facility is a different instrument.

How the financing could support Nebius’s expansion

Nebius intends to use the proceeds to accelerate the global build-out of its full-stack AI cloud. The company’s platform is designed to support workloads from data processing and model training through production deployment, so additional GPU capacity can support both AI-native companies and larger enterprise customers.

The practical benefit is timing. A cloud provider can have customer demand and contracts in place but still need to purchase GPUs, secure data-center capacity, fund networking and complete deployment before revenue is fully realized. Asset-backed debt can bridge that gap by turning an already operating infrastructure asset into growth capital.

Nebius says the current structure provides a framework for financing other long-term customer deployments. That statement is a company expectation, not a guarantee that future facilities will have the same pricing, collateral coverage or lender demand.

The broader infrastructure context is also important. Nebius and Nvidia announced a partnership in March to scale a full-stack AI cloud and enable deployment of more than five gigawatts of Nvidia systems by the end of 2030, according to an SEC-filed announcement. The July debt facility provides a financing mechanism for part of the capital intensity implied by that expansion, but it does not by itself prove that every planned deployment will be funded or completed on schedule.

Why the $40 billion commitment figure is important

Nebius says it already has more than $40 billion of additional contracted revenue from investment-grade customers, including Microsoft and Meta, and expects the financing approach to support further capital raising against other deployments.

The important distinction is between contracted revenue and cash already received. A customer commitment can improve visibility, but the company still has to deliver capacity, meet contractual milestones, operate the hardware and manage costs over the life of the agreement.

For investors, the useful question is not simply whether the commitment figure is large. It is whether the contracts produce predictable cash flows after deployment costs, financing expense, power, colocation, maintenance and hardware replacement are considered.

A repeatable model would require several conditions to remain true:

• Customer contracts must be sufficiently long and creditworthy for lenders to rely on them.
• GPU infrastructure must remain productive and valuable for the duration of the financing.
• Deployment schedules must be achieved without large cost overruns.
• Operating cash flow must cover debt service while still funding platform development.

Which banks participated in the transaction

MUFG led the facility as structuring agent, sole bookrunner and underwriter. Nebius also listed ABN AMRO, Bank of America, Deutsche Bank and HSBC as mandated lead arrangers; Citi, Crédit Agricole CIB, ING and Morgan Stanley acted as senior lead arrangers, with Goldman Sachs participating in the syndicate.

The transaction was described as significantly oversubscribed in Nebius’s announcement. That is evidence of strong lender interest in this specific financing, but it should not be read as proof that all future AI infrastructure debt will be equally available or equally priced.

Morgan Lewis separately confirmed that it advised Nebius on the transaction in its July 20 legal transaction notice. The firm described the facility as Nebius’s first senior secured debt facility backed by GPU infrastructure and contracted cash flows.

What Nvidia’s 9.3% stake changes for the market narrative

Nvidia’s disclosure added a second catalyst to the financing story. Investing.com reported that Nvidia disclosed a 9.3% passive ownership stake in Nebius through an SEC Schedule 13G filing dated July 13 and made public on July 20, with approximately 22.26 million shares counted in the beneficial ownership calculation.

The reported holding includes roughly 1.19 million existing shares and approximately 21 million shares associated with the warrant from Nvidia’s previously announced strategic investment. The same report said the warrant could not be exercised before September 11, 2026, even though the shares were included in the beneficial ownership calculation.

For market participants, Nvidia’s involvement can be interpreted as strategic validation of Nebius’s role in the AI infrastructure ecosystem. That is an investor interpretation, not evidence that Nvidia guarantees Nebius’s revenue, margins or share price.

The disclosure also creates an important analytical trap. A headline may combine the $775 million debt facility, Nvidia’s ownership disclosure and earlier strategic financing into one bullish narrative, but each event has a different effect on capital structure. Debt increases obligations, the warrant creates potential equity exposure, and the 9.3% disclosure primarily changes visibility around Nvidia’s economic interest.

The main risks behind the asset-backed model

The financing reduces reliance on immediate equity issuance, but it does not remove the risks of building AI data-center capacity. Nebius itself warned that future performance depends on its ability to build, operate and manage the business at scale, secure and retain customers, obtain additional capital and manage competitive and pricing pressures.

GPU economics can also change quickly. Hardware may depreciate faster than expected as new accelerator generations arrive, while electricity, networking, cooling and colocation costs can affect the margin available to service debt. A facility secured against deployed GPUs is therefore only as strong as the utilization and cash generation of those assets.

Customer concentration deserves separate attention. Investment-grade counterparties can lower credit risk, but dependence on a small number of large contracts can still expose the company to implementation delays, renegotiations or changes in capacity requirements.

Typical mistakes in evaluating the announcement include:

• Calling the transaction equity-free without considering Nvidia’s separate warrant-related arrangement.
• Counting contracted revenue as if it were current operating cash flow.
• Assuming the first facility automatically guarantees access to tens of billions in future debt.
• Ignoring the floating-rate exposure created by SOFR plus 2.50% pricing.

What investors should monitor after the announcement

The next step is to test whether the financing structure translates into operating execution. Investors should review Nebius’s future reports for disclosures about debt balances, collateral, interest expense, customer concentration, GPU deployment milestones and the conversion of contracted commitments into recognized revenue and cash flow.

A practical monitoring framework can be organized around four questions:

1. Are new GPU deployments delivered on the contractual schedule?
2. Does utilization rise quickly enough to support debt service and acceptable margins?
3. Does Nebius add debt faster than it adds durable cash-generating capacity?
4. Are new facilities financed at terms comparable to the inaugural transaction?

It is also worth comparing financing with deployment rather than looking only at headline capital raised. The company may have access to substantial commitments, but the economic result depends on the cost of each deployment and the revenue and margin generated once the infrastructure is in service.

What the deal means for AI infrastructure finance

Nebius’s transaction illustrates why AI cloud infrastructure is moving beyond a simple equity-funded growth model. When a provider has deployed GPUs, identifiable customer contracts and predictable cash flows, lenders may be willing to finance assets directly instead of requiring the company to fund every expansion through shareholders.

That model can be attractive for companies operating in capital-intensive markets, but it is not universal. It works best when customer commitments are credible, collateral is operational and the revenue profile is sufficiently stable to withstand hardware and market cycles.

As of July 21, 2026, the most defensible conclusion is that Nebius has secured a meaningful non-dilutive funding tool and demonstrated lender appetite for its first secured facility. The harder test will be whether the company can repeat the structure while expanding capacity, controlling leverage and converting its contracted demand into durable free cash flow.

For investors evaluating the story, the next action is to track execution against those measures rather than treating the $775 million headline or Nvidia’s 9.3% stake as a standalone investment case. The financing improves Nebius’s options; it does not eliminate the operational and balance-sheet discipline required to use them well.

Also read:

• Alpaca Raises $135 Million Equity for Tokenized Markets Infrastructure
• TerraFirma Raises $115 Million for Infrastructure Construction Technology
• Meta's Alberta Data Center Accelerates the AI Infrastructure Race
• Kling AI Raises $2.8 Billion at $18 Billion Valuation in July 2026 Round