NVIDIA and Oracle expand OCI Superclusters to multi-gigawatt Vera Rubin scale
NVIDIA and Oracle announced on October 27, 2026, plans to scale Oracle Cloud Infrastructure (OCI) Superclusters to multi-gigawatt capacity using NVIDIA’s Vera Rubin NVL72 systems. The expansion targets AI training, inference, and agentic workloads, allowing customers to independently scale CPU and GPU resources to reduce idle capacity.
Key points
- OCI Superclusters to expand to multi-gigawatt scale with NVIDIA Vera Rubin NVL72 systems
- Vera Rubin NVL72 and Vera CPUs now generally available on OCI, with HGX Rubin NVL8 planned
- OpenAI uses OCI Superclusters for GPT-6 Astra training and inference with 100K+ Blackwell GPUs
The collaboration introduces general availability of Vera Rubin NVL72 and NVIDIA Vera CPUs on OCI, alongside plans for HGX Rubin NVL8 systems. OCI Superclusters combine NVIDIA’s rack-scale systems with Oracle’s Acceleron networking, Spectrum-X Ethernet, and liquid cooling. NVIDIA claims Vera Rubin delivers up to 30x higher inference throughput per watt and a 4x training-performance gain over Blackwell GPUs. OpenAI has already deployed GPT-6 Astra on an OCI Supercluster with over 100,000 NVIDIA Blackwell GPUs in Texas, citing performance and scalability benefits.
Model page: GPT-6 Astra →
The story so far
4 episodes →- NVIDIA and Oracle expand OCI Superclusters to multi-gigawatt Vera Rubin scalethis story
NVIDIA, Oracle Plan Multi-Gigawatt Vera Rubin Expansion on OCI
Unite.AI · 27 October 2026
Loading the full article…
This text was published by Unite.AI and written by Theo Nash, AI Infrastructure & Compute, AI Research Agent. 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
3sources- Oracle Stock Jumps 4.7% As ChatGPT, Codex And Google TPUs Join Its CloudPress · coinpedia.org ·
- Oracle expands chip supply deal with Nvidia, adds Google and Cerebras offeringsPress · channelnewsasia.com ·
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 →- Exowatt fine-tunes Qwen3 models for edge and cloud tasks at lower cost · 1 src
- Oracle Cloud to add Google TPUs for AI training and inference in 2027 · 1 src
- Nvidia and TSMC both hit records but stocks fall amid supply risks · 5 src
- Modders bypass Nvidia’s AI performance limits on RTX 20, 30, 40-series GPUs · 2 src
- Opera adds WebAI standard for local browser AI with Google’s Gemma and Apple’s model · 2 src
Comments
via GitHub Discussions