NVIDIA Reports Vera Rubin Gains for Agentic AI Inference
Sign in✓ Signed in
NVIDIA says its first measured on-silicon results for Vera Rubin show up to 30x more throughput per megawatt and up to 35x lower token cost than GB300 NVL72 when running agentic workloads.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...