Vera Arrives: NVIDIA’s AI-Native CPU Enters the Global Supply Chain
NVIDIA has begun shipping Vera, its first CPU specifically designed for AI agents. The chip architecture optimizes for the high-throughput, trillion-parameter workloads of modern AI factories.
The semiconductor landscape is undergoing a tectonic shift as compute requirements move from general-purpose processing to AI-native architectures. NVIDIA has officially begun shipping its 'Vera' CPU, the company's first processor designed from the ground up to support autonomous AI agents. Vice President Ian Buck personally delivered the first systems, marking the start of a new era for hyperscale data centers.
Vera is not a traditional CPU; it is an 'XPU' designed to sit at the heart of an AI factory. Its architecture is optimized for 'tokens per second' and 'tokens per watt,' the new metrics of the generative AI era. By tightly integrating with NVIDIA’s NVLink Fusion and HBM (High-Bandwidth Memory) systems, Vera eliminates the bottlenecks that typically slow down trillion-parameter models. The focus here is on inference and agentic behavior—the ability for AI to process, reason, and act in real-time.
This release signals NVIDIA's intent to own the entire stack of the AI data center. As AI agents become more autonomous, the demand for specialized silicon that can handle complex logic alongside massive parallel processing will only grow. Vera is the opening salvo in a new chip war where the winner is determined by how efficiently they can turn electricity into intelligence.
Source: NVIDIA Blog
