Reported by 2 outlets
The story so far
AI summary of the reporting below · Updated · How we work
Nvidia is considering a recurring software fee of roughly $2,000 per graphics chip each year to operate artificial intelligence agents in data centers. Chief Executive Jensen Huang floated the potential licensing model at the GTC Berlin conference on October 22, framing it as part of an effort to expand software into a pillar of the company's business. Nvidia has not yet announced a formal launch date or confirmed final pricing. [1][2]
The proposed subscription would support workloads for autonomous agents, which execute tasks, tap enterprise data, and call external tools with minimal human supervision. Just before Huang's remarks, Nvidia unveiled its OpenShell Reference Architecture, which pairs Vera Rubin computing systems, Vera processors, and BlueField-4 data processing units running security software to quarantine misbehaving models. The company already sells its AI Enterprise platform for $4,500 per chip annually, and it remains unclear how the new tier would interact with existing licenses. [1][2]
The software strategy highlights Huang's ambition to transform the semiconductor giant into a full-stack platform provider. Outside partners are already adopting the surrounding toolset; Anthropic is pairing OpenShell with Claude Managed Agents, while Salesforce built a Slack bridge to manage permissions. Analysts note that attaching a $2,000 annual license to five million chips could yield $10 billion in recurring revenue every year. [1][2]
Nvidia is considering a $2,000 annual software fee per GPU for running AI agents. The proposed licensing model could generate billions in recurring revenue, adding to the company's existing AI software business, although pricing and rollout details remain unclear.
via Notebookcheck
Nvidia CEO Jensen Huang is reportedly weighing annual software fees of around $2,000 per GPU for enterprise AI, a model that could generate $10 billion a year if five million licenses were sold. The pricing and launch timeline remain unconfirmed, while Nvidia continues to expand its AI software and security tools for enterprise customers.
via Tom’s Hardware