The addition of Vera Rubin to Superclusters builds on Oracle's existing Nvidia Grace Blackwell compute, which was used by OpenAI at a facility in Abilene, Texas, to train its GPT-6 Astra model. OpenAI utilized an OCI Supercluster comprising more than 100,000 Blackwell GPUs for that training run. "As our frontier models advance, we need infrastructure that can keep pace," said Sreekanth Pothanis, engineering leader for compute infrastructure at OpenAI. "Nvidia Vera Rubin on OCI gives us the computing performance and scale to accelerate model training and serve increasingly demanding AI workloads."
The deployment of Vera CPUs specifically targets agentic AI workloads, combining high single-thread performance with substantial memory bandwidth to help AI agents complete CPU-intensive tasks more quickly. "Oracle Cloud Infrastructure gives customers the performance, security and scale to advance frontier AI," said Mahesh Thiagarajan, president of Oracle Cloud Infrastructure. "Our collaboration with Nvidia extends those capabilities across training, inference and agentic AI. By combining Nvidia Vera Rubin GPUs and Vera CPUs with Oracle Acceleron networking and AI-optimized data centers, we're helping customers keep GPUs productive, secure, and scale efficiently as their needs grow."
Beyond Nvidia hardware, Oracle announced a multi-year collaboration with Intel to expand Intel Xeon-based computing on OCI, including current Xeon platforms and planned OCI X14 offerings based on Intel's next-generation server processor codenamed Diamond Rapids. The partnership will increase CPU capacity on OCI while giving customers access to familiar tooling and software environments. "As AI agents scale, they need processors that can run software, move and secure data, and accelerate inference," said Kevork Kechichian, executive vice president and general manager of Intel's Data Center Group. "Xeon brings together performance and hardware-based confidential computing, giving organizations a trusted, cost-effective foundation for AI at scale." The planned Diamond Rapids-based OCI X14 offerings will target both AI and general-purpose computing, including Intel Advanced Matrix Extensions with FP8, FP16 and BF16 support for inference, and Intel Advanced Performance Extensions to help applications execute with less instruction overhead.
Oracle is also nearing general availability of its E7 instances powered by sixth-generation AMD Epyc processors, with AMD Pensando DPUs supporting Oracle Acceleron NICs for AMD-powered instances. The E7 instances use Epyc 9006 Series processors providing up to 256 cores and 512 threads per socket, up to 1,024 MB of L3 cache, 16 memory channels with up to 8000 MTps, and PCIe 6.0 connectivity, delivering significant uplifts over fifth-generation Epyc CPUs for enterprise workloads and agent deployment per instance and per watt.
Oracle further revealed plans to bring a variety of chip architectures onto its platform, including Cerebras' wafer-scale compute, SambaNova's SN50 chip, and the Positron Atlas chip, which is now generally available on OCI. Positron Atlas Inference Servers are purpose-built for large language models and deliver low-latency performance for popular open-weight models, networked through Oracle Acceleron. Atlas is Positron's first-generation chip, fabricated by Intel in the United States, and the company claims the hardware can achieve three times the compute-per-watt of Nvidia's H100 GPUs and support up to 500-billion-parameter models. Cerebras Inference on OCI is planned to be available in calendar year 2027, with Andrew Feldman, co-founder and CEO of Cerebras, noting that "every step an agent makes requires inference, whether it's reasoning, checking its work, or taking action. The faster the inference, the more work AI can do without making users wait." Details about the SambaNova partnership remain sparse beyond the planned SN50 deployment, with no specific timeline provided. These new chip offerings join the Google TPU, which Oracle revealed is set to be deployed on OCI under a multi-year, multi-gigawatt contract.
The broadened silicon strategy underscores Oracle's ambition to position OCI as a neutral, multi-architecture platform for AI training and inference, giving enterprises and AI labs the flexibility to match workloads to the most suitable processors. By offering Nvidia, Intel, AMD, Google, Cerebras, SambaNova and Positron hardware under a single cloud umbrella, Oracle is betting that customers will prioritize performance and supply diversity over single-vendor lock-in as AI compute demand continues to surge.