aiqbench analysis
NVIDIA's Vera Rubin architecture demonstrates a significant leap in energy efficiency for agentic AI workloads, achieving up to 30x higher throughput per megawatt compared to the previous GB300 NVL72 generation. This performance gain is critical as the industry shifts toward scaling complex agent systems that demand substantial computational resources and power management.
NVIDIA’s Vera Rubin delivers 30x more AI work per megawatt
NVIDIA Vera Rubin and Blackwell raise the bar for AI performance, delivering up to 30x more AI-factory throughput per megawatt than GB300 NVL72.
NVIDIA’s Vera Rubin NVL72 preview results achieve up to 30 times higher AI-factory throughput per megawatt than GB300 NVL72, according to testing with the SemiAnalysis AgentX benchmark. This leap in performance arrives as the scale of agentic AI rapidly expands; OpenRouter’s State of AI report found single agentic requests now consume 15 times the tokens of ordinary chat. AgentX specifically evaluates AI infrastructure for agentic-coding inference by replaying…
Excerpt shown under the publisher's RSS terms · 68 words · full text and figures are on the publisher's site
Related coverage
NVIDIA Vera Rubin Driving Performance Per Watt, Lowest Token Cost for Partners Worldwide
NVIDIA · Jul 21, 2026
NVIDIA Vera Rubin and Blackwell Set a New Standard for Agentic AI Performance per Watt
NVIDIA · Aug 24, 2026
AI Infra Summit: NVIDIA Vera Rubin and DSX Platform Advancements Showcase Energy Efficiencies of Optimizing Tokens Per Watt for AI Factories
NVIDIA · Sep 15, 2026
Up to 30x More Work Per Watt: NVIDIA Vera Rubin NVL72 Sets a New Efficiency Standard for AI Agents
NVIDIA · Aug 24, 2026