> ## Content Index
> Fetch the complete content index at: https://bytevyte.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Blue Owl Leads $2.4B GPU Equipment Financing for IREN
- URL: https://bytevyte.com/blue-owl-leads-2-4b-gpu-equipment-financing-for-iren/
- Published: 2026-08-29T12:41:56.000Z
- Updated: 2026-08-29T12:41:56.000Z
- Description: Blue Owl leads a $2.4B GPU equipment financing for IREN's Mackenzie campus, funding Blackwell Ultra GPUs at 9%.
- Author: Bytevyte Editorial
- Tags: ai-beats

Blue Owl Capital is leading a $2.4 billion GPU equipment financing package for IREN Limited's Mackenzie data center campus in British Columbia. Announced August 28, 2026, the package is split into two $1.2 billion pieces: a senior secured term loan and senior secured notes carrying a fixed 9.0% coupon. It covers roughly 90% of the GPU capital expenditure tied to the site, with the proceeds paying for NVIDIA Blackwell Ultra GPUs.

Pacific Investment Management Co. co-participates in the financing, which ranks among the largest GPU-backed debt transactions announced so far. The debt is secured against the hardware, and repayment is scheduled around GPU delivery and installation rather than a single upfront drawdown. Because the term loan is delayed-draw, Sydney-based IREN can fund purchases on a rolling basis within a set window. The fixed coupon on the notes gives IREN predictable interest costs across the life of the deal.

IREN is an AI cloud provider that owns data centers, compute capacity, and software for AI training and inference workloads. The NASDAQ-listed company recently delivered a 50MW liquid-cooled deployment to Microsoft in Texas, and its global development pipeline now exceeds 5GW of data center capacity. The Mackenzie expansion is air-cooled, putting IREN in a position of running both cooling approaches in parallel as it scales.

IREN's financing strategy splits by customer type. The $3.6 billion investment-grade facility is tied to the Microsoft contract, while the $2.8 billion tranche backs what IREN's management describes as the rest of the AI ecosystem, meaning cloud customers without investment-grade ratings. That separation lets lenders price each pool of hardware against the demand contract behind it, which is why the same company can borrow at 6.0% and 9.0% within the same quarter. The split shows how GPU equipment financing is priced against the credit quality of the underlying customer contract.

## Inside the GPU Equipment Financing Structure

The money supports the Mackenzie site's air-cooled NVIDIA Accelerated Computing Infrastructure, which includes Blackwell Ultra GPUs. Blue Owl describes the transaction as AI factory financing, a framing that treats GPU clusters as production assets with a defined output and a recoverable value. Delivery-linked repayment matters because GPU lead times and installation windows vary; tying repayments to when hardware enters service reduces the chance that a borrower carries debt on equipment that is not yet generating revenue.

The 9.0% coupon reveals how lenders price AI infrastructure credit today. IREN's earlier $3.6 billion investment-grade financing for its Microsoft contract carried a 6.0% rate and, combined with customer prepayments, funded 96% of the associated GPU capex. The Mackenzie facility costs 9.0% and funds about 90% of capex.

| Financing                             | Amount | Fixed rate | GPU capex funded     |
| ------------------------------------- | ------ | ---------- | -------------------- |
| Microsoft contract (investment grade) | $3.6B  | 6.0%       | 96% with prepayments |
| Mackenzie expansion (Blue Owl, PIMCO) | $2.4B  | 9.0%       | \~90%                |

That 300-basis-point spread is a direct measure of capital-access bifurcation in the AI cloud market. Operators who attach investment-grade customers to their capacity borrow at roughly two-thirds the cost of peers building on speculative demand. For IREN, the 9.0% rate is the price of speed: the company is racing to build AI cloud capacity in Canada while its customer base is still forming. For NVIDIA, the financing de-risks a large hardware order, since the purchase is backed by committed capital rather than IREN's cash flow alone.

The 90% coverage rate leaves IREN to fund the remaining tenth of GPU capex from its own resources, including customer prepayments and balance sheet cash. That split mirrors the Microsoft deal, where prepayments lift funding coverage to 96%. In both cases, the company retains a stake in the hardware it buys rather than offloading the full cost to lenders.

## IREN's Capital-Raising Sprint

IREN has secured $6.5 billion in GPU financing within three months across both investment-grade and non-investment-grade customer deployments, according to its fiscal 2026 results. The latest $2.8 billion tranche for non-investment-grade work includes the Blue Owl-led GPU equipment financing for the Mackenzie campus. The company's AI cloud services revenue more than doubled quarter over quarter to about $70.5 million in its most recent report, supported by $2.8 billion in new AI cloud contracts, and its results show AI compute pricing up 125%.

That scale puts the financing far ahead of current revenue, a reminder that IREN's capacity spending is running well beyond today's sales. The pace of borrowing reflects a strategic pivot. IREN has moved from Bitcoin mining into AI cloud services, and it now directs its balance sheet almost entirely toward compute infrastructure, with the 5GW-plus pipeline providing a long runway of construction ahead. The 9.0% coupon on the Mackenzie money is the cost of that trajectory for a borrower still building its customer base, and it is the clearest sign yet that GPU debt has become a mainstream financing tool for AI cloud operators.

For Blue Owl and PIMCO, the deal extends a pattern in which private credit managers underwrite GPU fleets as a distinct asset class. Asset-backed structures with delivery-linked drawdowns give capital providers a recoverable claim on the hardware itself, which is what makes large-scale financing of a non-investment-grade operator possible. The transaction also illustrates how the AI buildout is shifting from equity toward debt: a company that began as a Bitcoin miner can draw billions in institutional credit for hardware within a single quarter.

The participation of two large asset managers points to growing competition in private credit for AI infrastructure. GPU-backed loans give managers a route into the sector at fixed yields, and IREN's ability to close $6.5 billion in a single quarter suggests lender appetite is not the constraint on this buildout. At 9.0%, the notes give private credit investors a fixed yield secured against hardware, a structure that is easier to underwrite than unsecured corporate debt.

## Why this matters

The Mackenzie financing sets a reference point for the wider AI infrastructure market: GPU equipment financing now has a clear rate card, with 9.0% for non-investment-grade capacity and 6.0% for hyperscaler-backed projects. For decision-makers, the takeaway is structural. The spread between those rates determines who can afford to fund Blackwell-class capacity, and it will shape how quickly non-investment-grade operators can bring new AI compute online. For investors, the gap between the 6.0% and 9.0% rates separates the sector's cheapest capacity from its most expensive.

## Related Articles

- [NVIDIA Blackwell Ultra Drives Shift Toward Industrial AI Factories](https://bytevyte.com/nvidia-blackwell-ultra-drives-shift-toward-industrial-ai-factories/)
- [Nvidia Lancium Investment Moves the AI Race onto the Grid](https://bytevyte.com/nvidia-lancium-investment-moves-the-ai-race-onto-the-grid/)
- [Valar Atomics $1B Raise Fuels Nuclear Factory Push](https://bytevyte.com/valar-atomics-1b-raise-fuels-nuclear-factory-push/)

✔Human Verified

---

*Researched and cross-referenced against primary sources by the Bytevyte editorial team. This article was generated with the assistance of artificial intelligence and reviewed by the Bytevyte editorial team.*