Neocloud providers Sharon AI and Lambda have separately closed GPU-secured debt facilities, adding $1.365 billion in combined financing to a sector that increasingly treats high-end accelerators and customer contracts as bankable collateral.

Sharon AI closes $365m SPV facility at 9.95 percent

Australia's Sharon AI has secured a $365 million senior secured, GPU-backed SPV debt facility priced at a fixed 9.95 percent. The company secured the facility against its GPUs and the associated cash flows they generate.

The deal marks the first in an expected series of GPU financings designed to fund Sharon AI's planned deployment of 68,000 Nvidia GPUs by mid-2027. Together with previous raises, the facility brings Sharon AI's total debt and equity funding to $2.6 billion over the past 10 months.

"As demand for sovereign and secure, trusted AI infrastructure continues to outpace available supply globally, and particularly across Australia, New Zealand and the broader Asia-Pacific, access to scalable debt capital is an important enabler of our growth," said James Manning, co-founder and CEO of Sharon AI.

"This facility demonstrates how we expect to access debt markets to fund our GPU deployments, leveraging our book of quality customer offtake now standing at a TCV of over $8.8bn," Manning continued. "This is designed to enhance return on equity and ultimately drive increased long-term shareholder value. With a strong balance sheet, growing contracted capacity pipeline and a disciplined approach to capital allocation, we believe we are well positioned to continue scaling our AI platform across the Asia-Pacific region."

Jarden Australia acted as sole financial advisor and arranger on the transaction.

Sharon AI recently signed an agreement to lease capacity at GreenSquare's Sydney data center, where it plans to deploy 8,200 Nvidia Blackwell Ultra GPUs.

Lambda raises $1bn at 6.78 percent fixed

Lambda has secured $1 billion in senior secured fixed-rate financing. The facility takes the form of an investment-grade, delayed-draw term loan that was marketed to insurance companies and fixed-income investors.

The company will use the proceeds to purchase and develop GPU cloud infrastructure for three customer deployments it is contracted to support. The loan carries a 6.78 percent fixed interest rate.

This is Lambda's second institutional credit facility, following a $926 million facility secured in August 2026.

"The capital in this offering underwrites infrastructure in decades, not quarters, and has funded us as a private company on the strength of our customer contracts," said Michel Combes, CEO of Lambda. "Building on our investment-grade Term Loan B and bank lending facility, this is the third new credit market Lambda has opened in the last 18 months. Our progression into deeper and more diversified pools of capital is the market's verdict on the durability of our contracts and the scalability of our business."

J.P. Morgan acted as sole coordinating lead arranger, structuring agent, and bookrunner.

GPU collateral gains traction among lenders

GPU-backed loans remain a relatively new financing instrument, and both deals illustrate how quickly the structure is maturing. Hardware has historically depreciated in value over time, which made it an unreliable asset against which to secure debt.

That calculation is shifting. Sustained demand for AI compute means even older hardware retains its value for longer, and lenders now accept GPUs — paired with large customer commitments — as credible collateral. CoreWeave and Nscale are among the other neoclouds that have secured similar loans.

The pricing gap between the two facilities is notable. Lambda's investment-grade structure, sold to insurance companies and fixed-income investors, came in at 6.78 percent fixed. Sharon AI's SPV structure, priced at 9.95 percent fixed, reflects both the company's earlier stage of institutional access and the Australian market context.

Both deals point to the same underlying dynamic: contracted AI compute capacity has become an asset class in its own right, capable of supporting debt raised across private credit, insurance capital, and syndicated loan markets. Sharon AI's $8.8 billion total contract value book and Lambda's three contracted customer deployments gave lenders revenue visibility that raw hardware alone could never provide.

For neoclouds, the appeal is straightforward. Debt secured against GPUs and their cash flows allows providers to fund accelerator purchases without further equity dilution — a capital structure question that grows more pressing as deployment targets run into the tens of thousands of units. Sharon AI's 68,000-GPU target by mid-2027 sits firmly in that category.

The two transactions also extend a broader trend of private AI infrastructure companies accessing institutional debt markets. Lambda has now opened three new credit markets in 18 months, moving from bank lending through Term Loan B to insurance and fixed-income investors. Its ability to raise as a private company, on the strength of customer contracts alone, signals how lenders now evaluate AI infrastructure risk.

Whether older GPUs continue to hold residual value will determine the durability of this financing model. For now, lenders are underwriting the hardware as if the AI demand cycle will outlast the depreciation curve.