Nvidia launches a massive program that fundamentally changes the approach to monetizing computing power. The tech giant is now ready to provide its resources to startups on a partial payment basis — in exchange for future revenue.
This move marks a shift from the traditional model of one-time GPU sales to a system of recurring payments. Nvidia is positioning itself not just as a component supplier, but as a strategic investor in the artificial intelligence ecosystem.
How does the new scheme work?
The essence of the model is simple: cloud providers purchase Nvidia equipment, then lease the computing power to startups that lack the resources to build their own data centers. As a result, Nvidia receives a fixed share of the profit generated by its processors at end customers.
"This scheme gives Nvidia a recurring income that directly depends on the utilization rate of the equipment," the company's official blog notes.
The program is based on an updated AI platform. The offering is now becoming available to a wide range of companies, which is particularly relevant amid cost optimization by some major clients.
Big plans and fierce competition
Startups that choose this model remain tied to Nvidia's chips and software for years to come. For example, Sharon AI plans to install up to 40,000 Grace Blackwell GB300 chips under this program. And Firmus is building a 360-megawatt campus in Batam, Indonesia, which will house up to 170,000 graphics processors.
Locking clients into its own ecosystem is becoming critically important due to competitor activity. Chinese developers recently successfully trained a large-scale neural network without using American processors, while many global clients continue to actively test alternative solutions.
Another bet on the tech boom
Such financial schemes evoke associations with cross-financing mechanisms among independent experts. It is known that the corporation has committed to investing up to $100 billion in the OpenAI project, and also holds an approximately 7% stake in CoreWeave, a major equipment buyer.
According to Morgan Stanley estimates, the total spending of the world's largest IT corporations on developing artificial intelligence technologies will exceed $800 billion by 2026. By 2027, these investments could reach a record $1.1 trillion, comparable to the US defense budget.
Meanwhile, relative stability was observed in the stock market. During trading on July 2, the issuer's securities closed at $194.69 per share. The business's market capitalization approached the $4.8 trillion mark, although this result is still somewhat below the historical highs of the current year.
Analyst's view: Nvidia's new model is not just a way to make money, but a strategic maneuver to maintain dominance. Amid growing competition from Chinese chips and alternative architectures, tying startups to the ecosystem through financial commitments becomes a powerful loyalty tool. However, the question remains: how sustainable is this model if the return on AI investments begins to slow down?