# Nvidia gears up to smash CPU hegemony as Vera starts shipping > Source: > Published: 2026-10-12 15:00:28+00:00 Nvidia’s Vera CPU offerings have commenced shipping in “multiple form factors,” the company confirmed. Speaking ahead of the OCP Global Summit in San Jose, Joe DeLaere, Nvidia’s product lead for AI infrastructure solutions, said in a press briefing that dedicated Vera racks were “shipping now.” The Vera CPU rack platform leans on Nvidia’s MGX modular reference architecture and acts as a dedicated side-cabinet for handling agentic workloads. The “discrete system” placed next to an NVL72, as DeLaere described it, is separated from the core Vera Rubin platform, as the CPUs in an NVL72 are “dealing with other types of orchestration management,” leaving not enough cores to perform tasks needed in agentic workflows. DeLaere told press that Vera provides more than 2-fold performance on agentic workloads compared to the previous Grace generation, measured using the SPEC CPU 2026 benchmark. He also cited 1.8-fold performance on Terminal Bench, which he described as measuring agent tasks involving real code running in real containers. “If you get a Vera CPU rack, it's got over 20,000 cores available for agents to operate quickly and complete these tasks as fast as possible,” DeLaere explained during a briefing. “Because Vera CPU was designed specifically for agents with its high bandwidth and its high single-threaded performance, we see that it really accelerates the agentic tool calling and execution of code … and having those separate is really how we can achieve these tremendous agentic workload benchmarks.” Nvidia has been gearing up to take on the CPU hegemony of AMD and Intel at a time when the demand for agentic workloads has shaken up the processor space, with Arm even breaking its long-standing licensing wheelhouse to [launch physical chips](https://www.sdxcentral.com/news/arm-follows-nvidia-into-a-crowded-enterprise-cpu-race/). Nvidia first announced plans for a [standalone CPU business](https://www.datacenterdynamics.com/en/news/nvidia-to-make-vera-cpu-available-as-standalone-offering-with-coreweave-first-customer-to-get-access/) at the turn of the year, with the likes of [Meta](https://www.datacenterdynamics.com/en/news/meta-and-nvidia-sign-multi-year-partnership-for-cpu-gpu-and-networking-deployments/) and [CoreWeave](https://www.datacenterdynamics.com/en/news/coreweave-makes-nvidia-vera-rubin-nvl72-available-cognition-is-early-customer/) among early customers. [Amazon Web Services](https://www.aboutamazon.com/news/aws/aws-nvidia-2-million-gpus-ai) is also in line for Vera, with the hyperscaler billing it as an additional compute option for agentic workloads. Vera [entered full production](https://www.sdxcentral.com/news/nvidia-vera-cpu-enters-full-production-pitched-at-agentic-ai-workloads/) in the spring, with it set to replace Grace as Nvidia’s flagship processor. It also marks somewhat of a departure from Nvidia’s traditional reliance on Arm designs for CPUs in that it features a core design developed largely in-house, though it does still use the Arm v9.2-A architecture. In addition to availability as a standalone processor and a dedicated rack platform, Vera will also provide the CPU portion for the [Vera Rubin Superchip](https://www.sdxcentral.com/news/nvidia-announces-vera-rubin-superchip-for-late-2026/), also set to release in the latter half of this year.