Vera Rubin NVL72 inference tests show up to 7x better token throughput per MW vs. Blackwell on a 1.6T DeepSeek model, above Huang's 3x claim for 1T-3T LLMs (Bryan Shan/SemiAnalysis)
retained text
Techmeme permalink: https://www.techmeme.com/260914/p41#a260914p41 <a href="https://newsletter.semianalysis.com/p/vera-rubin-nvl72-agentic-inference"><img align="RIGHT" border="0" hspace="4" src="http://www.techmeme.com/260914/i41.jpg" vspace="4" /></a> <p><a href="https://www.techmeme.com/260914/p41#a260914p41" title="Techmeme permalink"><img height="12" src="http://www.techmeme.com/img/pml.png" style="border: none; padding: 0; margin: 0;" width="11" /></a> Bryan Shan / <a href="https://newsletter.semianalysis.com/">SemiAnalysis</a>:<br /> <span style="font-size: 1.3em;"><b><a href="https://newsletter.semianalysis.com/p/vera-rubin-nvl72-agentic-inference">Vera Rubin NVL72 inference tests show up to 7x better token throughput per MW vs. Blackwell on a 1.6T DeepSeek model, above Huang's 3x claim for 1T-3T LLMs</a></b></span> — Jensen Sandbagging Performance Again, 2x more Annual Profit Per GigaWatt, The More you Buy, The More you Earn, AgentX, InferenceX, Extreme Co-Design</p>
document fetched 2026-09-15 02:42