Optimizing AI Models – Interview with Quentin Anthony of Zyphra

Опубликовано 19 декабря 2024, 16:00

In this video, Ronak Shah, Product Marketing Manager for AMD Instinct™, and Quentin Anthony, Model Training Lead from Zyphra, discuss the latest advancements in AI model optimization.

Quentin details how Zyphra is revolutionizing personalized AI with user-specific weights for efficient model pipelines. He explains how hybrid state space models merge attention and Mamba blocks for high performance with lower memory and latency. This conversation also covers how AMD Instinct™ accelerators boost memory efficiency for real-time AI on devices like phones and AI PCs.


Access AMD ROCm™ platform developer resources: amd.com/en/developer/resources...

Learn about AMD Instinct™ accelerators: amd.com/en/products/accelerato...

Learn more about Zyphra training models: zyphra.com/post/the-zyphra-tra...

The information contained in this video represents the view of AMD or the third-party presenter as of the date presented. AMD and/or the third-party presenters have no obligation to update any forward-looking content in the above presentations. AMD is not responsible for the content of any third-party presentations and does not necessarily endorse the comments made therein. GD-84.

Find the resources you need to develop using AMD products: amd.com/en/developer.html

***

© 2024 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, EPYC, ROCm, and AMD Instinct and combinations thereof are trademarks of Advanced Micro Devices, Inc.