NVIDIA Vera Rubin NVL72 delivers 30x more agentic work per watt
NVIDIA's Vera Rubin NVL72 systems deliver up to 30x higher throughput per megawatt than GB300 NVL72 on agentic workloads, measured with SemiAnalysis AgentX. Agentic requests consume 15x more tokens than chat, and token cost is 35x lower.
How this story unfolded
4 days · 4 reports · 4 of 5 shown
- Aug 22
- Aug 24
- Aug 25
NVIDIA by email
Get an email when NVIDIA has news
No news that day, no email.
More stories today
- Stable Diffusion user tests H3 model with Cheers-style script
- Reddit users share impressive image-to-video AI demos
- Reddit reminds users they can legally seed AI models via torrenting
- MiniMax H3 reverse-engineers paintings into basic forms
- OpenAI DevDay Exchange Seoul applications close Sept 4