Chips and hardwareUpdated November 6

NVIDIA makes a speed-and-efficiency case for Vera

AI agents do not stay on one kind of processor: NVIDIA says they move from GPU-based reasoning to CPU tasks such as searching, querying databases, and writing or testing code. Vera is designed for that handoff. In a comparison framed around a 100-megawatt data center, NVIDIA claims 1.6x more agents per watt than the fastest x86 CPU, and 2x faster agent speed than the most energy-efficient x86 CPU. Separately, Wccftech’s headline says Vera is 2x faster than a 256-core AMD EPYC Venice and handles 1.6x more agents per watt than a frequency-optimized 96-core Venice.

2 publishers2 reports

Coverage started November 5