Efficient Inference on MI300X: Our Journey at Microsoft, Rajat Monga, Microsoft, CVP AI Frameworks

Опубликовано 18 декабря 2024, 17:34

In this Advancing AI 2024 Luminary Developer Keynote, Rajat Monga, CVP AI Frameworks at Microsoft, discusses efforts in deploying key models on AMD Instinct™ MI300X GPUs. Rajat starts with why they believed it was a good idea to try MI300X; he covers the inside story of what it took to bring up a model on a new machine, to driving performance optimizations that made it competitive against Nvidia H100.

Gain access to AMD developer tools and resources.
amd.com/en/developer.html#soft...

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.

© 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.