Nvidia Scores $500 Billion in Financing for Its Customers. That's A Lot.

I have often written about how Nvidia’s ability to use its size and resources to secure supply commitments for sought-after wafers, optical components, and memory chips gives it a significant advantage over its rivals.

Now, we can add a massive amount of AI infrastructure financing to the list.

On Monday, Nvidia announced six strategic partnerships with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR to form new AI compute infrastructure financing platforms that will raise over $500 billion of third-party capital.

“In AI, compute is revenue. NVIDIA compute is uniquely suited for this role. It is broadly adopted, flexible across models and workloads, fungible and transferable across customers and operators, and continuously improved through CUDA software — extending its useful life and improving its economics over time,” Nvidia CEO Jensen Huang said in the news release.

Jensen cited how one-year Nvidia GPU H100 rental pricing rose from about $1.70 per GPU-hour in October 2025 to about $2.35 per GPU-hour in March 2026, and how six-year-old Ampere A100s are still in service and generating revenue today.

Jensen and executives from the six financial firms also appeared on CNBC after the deal was announced Monday afternoon. There were a number of fascinating tidbits from the segment.

Nvidia GPUs are an investable asset class:

-Jensen stated each gigawatt costs $50 billion to $60 billion. Nvidia GPUs are special because they run every AI model, have proven to have a long life and productive revenue generation, and have the broadest ecosystem. That is why the AI projects using Nvida GPUs can be financed.

-One can infer that AI ASICs from hyperscalers do not have these attributes compared to Nvidia GPUs.

Jensen on how AI demand is still overwhelming supply:

-“We’re going to be constrained for some time. Pretty much across the board. From chips to memories to packaging to systems, photonics, connectors, land power construction workers. The whole thing. The entire supply chain. Up and down.”

On worries about the profitability of AI lab customers:

-“I believe within months you’re gonna realize these companies are extremely profitable. These are the fastest growing technology companies in history. And the tokens they are generating are incredibly profitable.” They will be the “biggest IPOs in history.”

Goldman Sachs CEO David Solomon on available capital:

-$500 billion sounds like a lot, but there are $9 trillion in U.S. money market funds and more than $100 trillion in U.S. equities. He has a “deep belief” in this opportunity and Goldman brings its “extraordinary distribution network” to this deal.

BlackRock CEO Larry Fink:

-He said it is a very attractive opportunity with long-dated, long-term returns. They will be talking to pension funds.

-“We’re going to be raising quite a lot.” The U.S. alone needs to build 70 gigawatts and we “need to raise this money as fast as possible” to make the U.S. the leader in AI.

-It will create a huge number of jobs. “We’re going to have to raise trillions of dollars” in the coming years.

Blackstone executive Jon Gray:

-A sevenfold increase in demand for LLMs within his company during the last six months. The amount of compute is not keeping up.

The third-party aspect of the capital is important. The market is skittish about Nvidia leaning too heavily on its balance sheet to fund infrastructure. When The Wall Street Journal reported in late July that Nvidia was in talks to provide a $250 billion backstop for an OpenAI data center project, the news drove AI stocks and Nvidia shares significantly lower.

At one point during the CNBC interview, it sounded like Jensen implied Nvidia will not backstop OpenAI in the scale reported by The Wall Street Journal.

CNBC’s Becky Quick: “That brings up this important question. There’s been all these big numbers being thrown around. What you’re doing with SK. There was an Wall Street Journal story recently that suggested you’d be backstopping financing $250 billion for OpenAI plant in Ohio. That’s a lot of money. It’s a lot of thing to carry on your balance sheet. This is not that. Because this is not money Nvidia is backstopping.”

Jensen: “Those two things [SK and OpenAI] are not that either .. With respect to OpenAI. I won’t comment about rumors.”

Later in the evening, Nvidia did say it might provide a “residual-value support mechanism” for up to 25% of a project on a case-by-case basis in the partnerships with the financial firms announced today. “That support is limited, residual-value based and designed to complement — not replace — independent underwriting. This is substantially lower than other compute-financing arrangements,” the company said.

What does this news mean for AI trade investors?