Nvidia is ending its revenue-sharing programs with AI cloud providers. The model, launched on July 1, 2026, allowed companies to acquire Nvidia GPUs with financial support from the manufacturer in exchange for a percentage of the revenue generated by these chips. Participants included Sharon AI in Australia, which planned to deploy up to 40,000 GB300 GPUs, and Firmus Technologies in Indonesia, with up to 170,000 GPUs and expected revenues of $25 to $30 billion over six years. In the second quarter of fiscal year 2027, Nvidia’s non-hyperscale segment represented approximately 50% of data center revenue. This decision is a setback for cloud providers who will now need to secure capital through other means.
Source: Read the original article
