--- title: "Nvidia offers startups its AI chips in exchange for a share of their revenue" description: "Nvidia has begun offering some young cloud companies access to its sought-after AI chips without the full upfront cost — in return for a share of the revenue they earn. The arrangement deepens the chipmaker's grip on the AI build-out, and revives questions about the industry's increasingly circular flows of money." category: "Technology" category_url: https://newsparlor.com/category/technology author: "Daniel Morales" published: 2026-07-02T10:30:00.000Z updated: 2026-07-02T10:30:00.000Z canonical: https://newsparlor.com/article/nvidia-offers-startups-its-ai-chips-in-exchange-for-a-share-of-their-revenue tags: ["nvidia", "artificial-intelligence", "gpu", "cloud-computing", "startups"] --- # Nvidia offers startups its AI chips in exchange for a share of their revenue Nvidia has begun offering some young cloud companies access to its sought-after AI chips without the full upfront cost — in return for a share of the revenue they earn. The arrangement deepens the chipmaker's grip on the AI build-out, and revives questions about the industry's increasingly circular flows of money. Nvidia, the company whose chips power much of the artificial-intelligence boom, is trying a new way to keep that boom going: helping cash-strapped young cloud providers get its hardware without paying the whole bill upfront, in exchange for a slice of their future income. ## How the arrangement works Under the model, which Nvidia has described as a way to widen access to AI computing, the chipmaker deploys its graphics processing units (GPUs) with smaller "AI cloud" companies on a revenue-sharing and credit-support basis, rather than requiring them to buy the hardware outright, [according to Nvidia](https://blogs.nvidia.com/blog/nvidia-unlocks-ai-compute-at-scale-capital-partners-to-power-ai-infrastructure-buildout/). Those providers then rent the computing power to their own customers. The twist is that Nvidia gets paid twice: once on the hardware, as usual, and again through a share of the cloud revenue that the supported capacity generates, [Seeking Alpha reported](https://seekingalpha.com/news/4609514-nvidia-strikes-revenue-sharing-deals-with-sharon-ai-cloud-partners-to-scale-ai-factories). For a startup, the appeal is obvious: high-end AI chips can cost tens of thousands of dollars each, and building meaningful capacity can run into the billions, a barrier that shuts out all but the best-funded newcomers. ## The first partners Among the first companies named as taking part are Sharon AI, which is deploying tens of thousands of Nvidia's latest-generation chips, and Firmus, which is building a large AI data-center campus in Batam, Indonesia, that is projected to scale to as many as 170,000 GPUs. Such projects require enormous financing; Firmus has separately raised billions of dollars in debt to fund its build-out. ## Why Nvidia is doing it The move fits a strategy Nvidia has pursued as demand for AI computing has surged: nurturing a broad ecosystem of companies that buy and deploy its chips, and binding them more closely to its platform. By sharing the financial risk, Nvidia helps ensure that smaller providers can keep expanding — and keep buying Nvidia hardware rather than turning to rival chipmakers. For Nvidia, aligning its own returns with the growth of its customers is a way to lock in future demand at a time when the scale of AI infrastructure spending is unprecedented and, some fear, potentially ahead of the revenue to justify it. ## The concerns The arrangement has also drawn scrutiny. Because Nvidia is effectively helping to fund companies that then spend the money on Nvidia products, analysts have pointed to it as another example of the AI sector's increasingly "circular" financing — where chipmakers, cloud providers and AI developers invest in one another in ways that can inflate apparent demand and blur who is really taking on the risk. Critics warn that such entanglement can leave smaller providers heavily dependent on a single supplier that is also a financial stakeholder, potentially limiting their independence. Supporters counter that Nvidia is simply lowering a real barrier to entry and spreading the cost of an expensive build-out — and that similar risk-sharing is common when new infrastructure is being built quickly. Whether the revenue-share model becomes standard practice or stays a targeted tool for a handful of partners will depend on how the first projects perform, and on whether regulators take an interest in how tightly the AI supply chain is becoming interwoven. For now, it is one more sign of how central Nvidia has become — not just as a supplier to the AI industry, but as a financier of it. ## Sources - [NVIDIA Unlocks AI Compute at Scale, Inviting Partners to Power the AI Infrastructure Buildout](https://blogs.nvidia.com/blog/nvidia-unlocks-ai-compute-at-scale-capital-partners-to-power-ai-infrastructure-buildout/) - [Nvidia strikes revenue-sharing deals with Sharon AI, cloud partners to scale AI factories](https://seekingalpha.com/news/4609514-nvidia-strikes-revenue-sharing-deals-with-sharon-ai-cloud-partners-to-scale-ai-factories)