NVIDIA Launches Vera CPU Purpose-Built for Agentic AI Workloads

NVIDIA

Mar 19, 2026 · Updated Apr 25, 2026

NVIDIA launched the Vera CPU at GTC, its first processor designed specifically for agentic AI and reinforcement learning. It delivers 2x efficiency and 50% faster performance than traditional rack-scale CPUs, with major cloud providers already signed on.

NVIDIA Vera CPU runs on 88 custom-designed Olympus cores with LPDDR5X memory delivering up to 1.2 TB/s of bandwidth — twice the bandwidth at half the power of general-purpose CPUs. Each core handles two tasks via Spatial Multithreading, targeting the concurrent workloads agentic AI demands. A single rack fits 256 liquid-cooled Vera CPUs sustaining over 22,500 concurrent environments at full performance.

Vera represents a new class of CPU, with Jensen Huang noting the CPU is now driving the model rather than simply supporting it. It pairs with Rubin GPUs via NVLink C2C at 1.8 TB/s — 7x PCIe Gen 6 bandwidth. Cursor, Meta, Alibaba Cloud, CoreWeave, and Oracle Cloud Infrastructure are among early adopters, alongside national labs like Los Alamos and TACC.

Evaluate Vera for agentic inference and reinforcement learning workloads — 256 liquid-cooled CPUs per rack sustaining 22,500+ concurrent environments changes how you'd architect large-scale distributed RL training.

View the full update on nvidianews.nvidia.com

NVIDIA Newsroom

@nvidianewsroomMar 16

Agentic AI is only as fast as the CPU behind it. #NVIDIAGTC news: Introducing the NVIDIA Vera CPU, the first CPU purpose built for agentic AI and reinforcement learning. Delivering 2x efficiency and 50% faster performance than traditional rack scale CPUs. https://t.co/Dd76FDFlmR

View on X

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

See all AI news & updates from NVIDIA →

Keep reading

NVIDIA Launches Vera Rubin Platform With Seven Chips for AI Factories

NVIDIA announced the Vera Rubin platform at GTC, putting seven new chips into full production for large-scale AI infrastructure. The NVL72 rack trains mixture-of-experts models with one-fourth the GPUs compared with Blackwell while delivering 10x inference throughput per watt.

SambaNovaApr 15

SambaNova Launches SN50 RDU to Power High-Speed Agentic Inference

SambaNova's fifth-generation SN50 chip is designed specifically for agentic workflows, offering 5x faster speeds and 3x lower costs than traditional GPUs. This hardware shift addresses the latency bottlenecks of multi-step reasoning and tool use in enterprise AI systems.

Meta Taps Millions of AWS Graviton Cores to Scale Agentic AI

MetaApr 24

Meta Taps Millions of AWS Graviton Cores to Scale Agentic AI

Meta is deploying tens of millions of AWS Graviton cores to support the high-compute demands of its next-generation agentic AI systems. The partnership signals a shift toward CPU-intensive reasoning and planning workloads as autonomous agents move to global production scale.