Neural Magic Demo: Achieve GPU-Class Performance on CPUs with Drastic Cost Savings
Neural Magic is a software solution for deep learning acceleration, enabling data scientists to use ubiquitous and unconstrained CPU resources to achieve performance breakthroughs at scale, no expensive hardware required. The Neural Magic Inference Engine enables machine learning to be practically run in new places, on new kinds of workloads. It delivers state of art, GPU-class performance for the deep learning applications running on commodity CPUs. All with drastic cost savings.
This is big. And this video shows that it’s in fact possible to achieve GPU-class performance on commodity CPUs without making sacrifices, at fraction of the cost.
This video gives you an introduction to Neural Magic, a demo of Neural Magic in action, and steps to deliver GPU-class deep learning performance and better without the expense.
Date recorded: May 5, 2020
Presenter: Bryan House, Chief Commercial Officer, Neural Magic