Vera Rubin's first test

NVIDIA has published the first MLPerf benchmark results for its next-generation Vera Rubin platform. According to the company, the platform delivers leading performance in training, post-training and agentic inference.

Sweeping the training benchmarks

NVIDIA said it won every benchmark in MLPerf Training v6.1. MLPerf is an industry-standard suite that compares the performance of hardware and software systems on AI workloads.

Efficiency for agentic AI

The standout result from the company's first MLPerf Endpoints submission was more than 10x generational gains in throughput per watt for agentic AI. This metric measures how much work a system does per unit of energy, which is critical for data center operating costs.

Why it matters

Vera Rubin is positioned as the hardware family that will succeed NVIDIA's current-generation platform. Because MLPerf results are a reference point for companies training and running AI models, they can directly influence the competitive balance in the sector.