Skip to content
VECTOR WIREAI INTELLIGENCE
UTC
Home

Nvidia Rubin HPC GPU packs over 50,000 CUDA cores for scientific AI workloads

Nvidia's Rubin HPC GPU and accompanying software tools target scientific and engineering workloads with a core count exceeding its current flagship chip.

Nvidia introduced Rubin HPC, an upcoming graphics card with significantly more cores than the company's current flagship chip, alongside over a half-dozen software tools for scientists and engineers2. The GPU features more than 50,000 CUDA cores1. The offerings are designed to speed up scientific research across a range of fields. Nvidia's Vera Rubin architecture recently swept every MLPerf Training v6.1 benchmark[1].

Was this useful?

The day’s vector, in your inbox.

Vector Wire Daily: one dispatch each publishing day.