Nvidia launches revenue-sharing program for AI startups to access GPU compute power
Nvidia announced a new revenue-sharing program on July 2, 2026, allowing AI startups to access its GPU compute power in exchange for a slice of future profits. The program provides token credits to cloud-based AI firms, with initial partners Sharon AI (Australia) and Firmus Technologies (Singapore) deploying over 200,000 Nvidia GPUs, including a planned 360-megawatt data center in Batam, Indonesia. This model addresses GPU scarcity and helps cash-poor startups avoid upfront hardware costs, while Nvidia gains recurring revenue linked to usage.
Editorial summary awaiting refresh
Cross-source coverage
Wire timeline
Nvidia Doubles Down on AI Startups With a Historic Revenue Sharing Model
Nvidia has introduced a novel revenue-sharing financing model to help smaller AI cloud companies access its hardware without paying the full upfront cost. The chipmaker allows partners to build AI compute capacity and share future cloud revenue instead of requiring outright purchases. Sharon AI (Australia) will deploy up to 40,000 Nvidia Grace Blackwell GB300 GPUs, while Firmus is building a 360-megawatt campus in Indonesia with up to 170,000 GPUs. The model targets AI-native firms needing fast compute access for training and inference. Critics note potential circular financing risks, as Nvidia separately plans to raise $20 billion in debt. The initiative could reshape who can build next-generation AI infrastructure.
Yahoo FinanceNvidia Launches Cloud and Revenue-Sharing Program to Boost AI Startup Access to Compute Power
Nvidia Corp. announced a new initiative to provide AI startups with access to high-performance computing infrastructure through a revenue-sharing and credit-support model. Under the program, AI cloud providers will offer services powered by Nvidia technology, allowing Nvidia to profit from both hardware sales and a share of future cloud provider earnings. Initial partners include Sharon AI, which plans to deploy up to 40,000 Nvidia Grace Blackwell GB300 GPUs, and Firmus, which is developing a DSX AI factory campus in Indonesia. The move is part of Nvidia's broader strategy to expand its AI ecosystem, following over $40 billion in AI investments in early 2026, including a $30 billion stake in OpenAI. The initiative has sparked debate, with critics like Michael Burry calling the circular investments 'Fugazi,' while supporters like Daniel Newman praise it as prudent investing that enables Nvidia to grow and generate more cash flow.
Yahoo FinanceNvidia offers start-ups compute power in exchange for revenue share
Nvidia announced a new program allowing AI-focused start-ups to swap access to compute power for a share of future profits. The chipmaker will provide token credits to cloud-based AI firms, model builders, and enterprises, positioning itself as an intermediary for full-stack computing. Two initial Australian partners, Sharon AI and Firmus Technologies, will deploy over 200,000 Nvidia GPUs, including at a planned 360-megawatt data center in Batam, Indonesia. The move highlights the scarcity of GPU compute power, often compared to oil, and comes amid broader industry trends where AI firms like OpenAI have signed similar revenue- and equity-sharing deals with chipmakers. Nvidia also announced plans to raise at least $20 billion in debt for general corporate purposes.
US Top News and AnalysisNvidia offers start-up customers chance to swap compute power for revenue share
Nvidia announced a new program allowing AI-focused startups to exchange access to compute power for a share of future profits. The revenue-sharing agreements provide token credits for cloud-based AI firms and model builders. Two initial Australian partners, Sharon AI and Firmus Technologies, will offer access to over 200,000 Nvidia GPUs, including via a data center in Batam, Indonesia. The move highlights the critical scarcity of GPU compute power for AI startups, which has been compared to oil and tied to futures contracts. Nvidia positions itself as an intermediary, while other AI firms like OpenAI have also entered similar revenue and equity-sharing deals with chipmakers. Nvidia also recently announced plans to raise at least $20 billion in debt for general corporate purposes.
US Top News and AnalysisNvidia Offers Revenue-Sharing Model on AI Cloud Hardware Sales
Nvidia announced a new optional financing vehicle that allows it to collect revenue twice from the same silicon: once from hardware sales to partner AI clouds, and again as an ongoing percentage of the cloud revenue generated by that hardware. The model, described as a 'revenue-sharing and credit-support' approach, aims to help cash-poor AI startups access compute by trading future revenue for token credits today. First named partners are Australia's Sharon AI and Singapore-based Firmus Technologies. Sharon AI's six-year agreement covers 72 MW of Australian data center capacity scaling to 40,000 Grace Blackwell GB300 GPUs, while Firmus is building a 360 MW campus in Indonesia. The model inverts Nvidia's previous approach of direct cash investments in customers like OpenAI and xAI, instead extending credit support and collecting royalties tied to utilization rather than hardware sales. Revenue-split percentages remain undisclosed.
Latest from Tom's HardwareNvidia Launches Revenue-Sharing Model for AI Startups
Nvidia announced a revenue-sharing program on July 2, 2026, designed to give AI startups, model builders, and enterprises access to its computing infrastructure without requiring upfront hardware financing. Under the model, revenue from product sales and cloud operations flows back to Nvidia, creating what the company calls a 'recurring, usage-linked earnings stream.' Token credits are also offered to startups. Nvidia named two initial partners: Sharon AI, which will deploy up to 40,000 Grace Blackwell GB300 GPUs, and Firmus Technologies, planning a campus in Batam, Indonesia with up to 170,000 GPUs and 360 megawatts of power. The program targets AI-native firms like Baseten, Fireworks AI, and Together AI. Nvidia framed the move as a response to growing demand for production inference workloads and increasing competition from custom AI chips by cloud providers like Alphabet and Amazon. Nvidia stock was down 0.5% in premarket trading.
Yahoo FinanceNvidia offers start-ups compute power in exchange for future revenue share
Nvidia announced a new partnership program allowing fast-growing AI start-ups to swap access to compute power for a slice of future profits. The revenue-sharing agreements provide token credits for cloud-based AI firms, model builders, and other enterprises. Two initial partners, Australia-based Sharon AI and Singapore-based Firmus Technologies, will provide access to over 200,000 Nvidia GPUs, including via a planned data center in Batam, Indonesia. The move highlights the critical scarcity of GPU compute power for AI start-ups, with GPUs likened to oil. Nvidia positions itself as an intermediary, while AI firms increasingly use revenue and equity-sharing deals to address liquidity issues. Nvidia also recently announced plans to raise at least $20 billion in debt for general corporate purposes.
US Top News and Analysis