Buck explained how Vera will help.<\/p>\n
\u201cWhen AI models are posed a question, the answer, often, isn\u2019t already prepped and ready to go. \u201cThe models actually have to generate some Python code to arrive at the correct answer,\u201d Buck said. A task at which the Vera CPU excels. \u201cThat\u2019s why we are seeing the demand for CPUs skyrocket,\u201d Buck continued.<\/p>\n
A trend the OCI team was also witnessing.<\/p>\n
\u201cOCI plans to deploy hundreds of thousands of NVIDIA Vera CPUs beginning in 2026 because agentic AI demands sustained performance at massive scale,\u201d said Batta. \u201cVera\u2019s architecture is purpose-built for high-throughput reasoning workloads, delivering the efficiency, density and footprint OCI needs to power the next generation of enterprise AI.\u201d\u00a0<\/p>\n
OCI is the first cloud provider to deploy Vera at hyperscale. For enterprise customers, that means production-grade agentic AI infrastructure at a scale no other cloud provider can match today.<\/p>\n
The OCI team was eager to put Vera to work, offering their customers another system to customize and validate their agentic AIs\u00a0 and workloads, Miller said. \u201cI am really looking forward to the reaction of people who come through here, and working together to get the most from Vera,\u201d he said.<\/p>\n
What Vera Delivers<\/b><\/p>\n
Vera is part of NVIDIA\u2019s extreme co-design story, alongside the NVIDIA Rubin GPU, BlueField 4 DPU, Spectrum-X and MGX rack architecture.<\/p>\n
In addition to powering standalone CPU systems, Vera is the host processor for Vera Rubin NVL72 where it pairs via second-generation NVIDIA NVLink-C2C to a pair of Rubin GPUs.\u00a0<\/p>\n
Vera CPU at a glance<\/p>\n
What it is<\/strong> \u2014 NVIDIA\u2019s first custom CPU, designed for agentic AI<\/p>\n What it handles<\/strong> \u2014 Orchestration, tool-calling, RL workloads, data analytics, agent sandboxing, long-context state management<\/p>\n Who it\u2019s for<\/strong> \u2014 AI labs, cloud providers, and enterprises running agentic AI at scale<\/p>\n Core specs<\/strong> \u2014 88 custom Olympus cores, 1.2 TB\/s memory bandwidth, 50% faster per-core under full load<\/p>\n In these systems, Vera and Rubin share a unified memory architecture that keeps accelerated compute highly utilized.<\/p>\n Vera\u2019s fast CPU cores and interconnect handle orchestration, control, and data movement needed to feed GPUs at 2x the energy efficiency of traditional infrastructure.<\/p>\n The age of agentic AI has a purpose-built CPU, and its name is Vera.<\/p>\n