Achieve the World's Highest AI Compute Efficiency

Summary

While many AI chips claim hundreds of Tera Operations Per Second (TOPS), most of them achieve only 40% of the peak performance, leaving more than half dark (unused) silicon. AMD-Xilinx achieved the world's highest compute efficiency at 90%, becoming the first vendor to achieve near Zero Dark AI Silicon in modern AI benchmarks. In this webinar, we will cover two partner solutions that both run on the VCK5000 development card achieving unmatched performance over mainstream GPUs.

Join us and learn about:

• VCK5000 Development Card: Key features and benefits
• Zebra Software Solution: A software stack that computes any neural network on an FPGA faster, with lower power consumption, plug & play ease of use, and at lower cost.
• Aupera Video ML/AI Streaming Server Solution: An extensible plug and play software stack for video ML/AI applications which enables fast deployment on Xilinx FPGAS platforms achieving low latency and high throughput.

At the end of this webinar you will be equipped with the necessary information to compute AI inference at high performance with Mipsology and achieve full video processing ML inference pipeline for AI recognition with Aupera

Featured Speakers

Andy Luo

Director of Product Marketing for AI, AMD-Xilinx

Jaydev Shelat

Head of ML Solutions, Mipsology

Robert Lara

Sr. Director Marketing & Business Development, Mipsology

Ness Afsham

Chief Scientific Officer, Aupera