Nvidia proposes $2,000 annual fee per GPU for AI software
Nvidia CEO Jensen Huang suggested an annual software fee of $2,000 per GPU at GTC Berlin on October 22, but the company has not confirmed terms, scope, or timing. The fee would apply to software managing AI agents in Nvidia’s ‘AI factories,’ though it remains unclear whether it replaces or complements existing subscriptions like AI Enterprise, which costs $4,500 per GPU annually for one year or…
Key points
- Nvidia CEO Jensen Huang mentioned a potential $2,000 annual software fee per GPU at GTC Berlin on October 22
- OpenShell Reference Architecture launched October 21, combining AI infrastructure and security tools like Sentry
- Existing AI Enterprise subscription costs $4,500 per GPU annually, but new fee’s relationship to it is unclear
The proposal follows the October 21 launch of OpenShell Reference Architecture, a full-stack software stack for AI workloads. OpenShell includes Vera Rubin infrastructure, Vera CPUs, OpenShell runtime isolation, and BlueField-4 DPUs with Nvidia Sentry for agent monitoring. Whether the $2,000 fee covers OpenShell or other components is unspecified. Enterprises must wait for official terms to assess value, as overlapping charges risk confusion. Nvidia has not disclosed how many GPUs would qualify or how multi-die chips would be counted.
The story so far
2 episodes →- Nvidia proposes $2,000 annual fee per GPU for AI softwarethis story
Nvidia Already Makes Billions Selling AI Chips. Now It Wants $2,000 a Year for Each GPU
gadgetreview.com · 22 October 2026
Loading the full article…
This text was published by gadgetreview.com and written by https://www.facebook.com/iamrexedison/. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
8sources- Hacker News discussion · 38 pointsnews.ycombinator.com
- Reddit discussionreddit.com
- Nvidia AI Software GPU licensing fee weighed at $2,000 per GPU: See full detailsPress · geo.tv ·
- Nvidia's AI chips already make billions, now Jensen Huang wants $2,000 per GPU in annual software feesPress · techspot.com ·
- Nvidia Financial Strategy Faces Stress Test As AI Credit Markets TightenPress · finance.yahoo.com ·
- SMH, seriously? -->> NVIDIA Eyes Software Revenue With a $2,000 per GPU Yearly Figure From HuangPress · techpowerup.com ·
- Nvidia discusses $2,000-per-GPU annual software fee for running agents on AI infrastructurePress · Tom's Hardware ·
- Nvidia discusses $2,000-per-GPU annual fee for software that 'runs agents on AI factories'Press · finance.yahoo.com ·
- Nvidia discusses $2,000-per-GPU annual fee for running AI agentsPress · notebookcheck.net ·
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- Ggml-org merges CUDA optimizations for MoE gate/up matmuls and GLU in MMQ · 1 src
- Nvidia older chips prices rise, signals AI demand · 1 src
- Core 4 Solutions lists 300-rack AI supercomputer with 9,600 Nvidia A100 GPUs on eBay for $59.7M · 2 src
- Moore Threads reports Q3 revenue up 1,774% to ~$240M · 1 src
- Cowboy Space unveils Big Iron rocket to launch 20,000 AI data center satellites · 1 src
Comments
via GitHub Discussions