939: Mixture-of-Experts and State-Space Models on Edge Devices, with Tyler Cox and Shirish Gupta

State space models (SSMs), granite models, and Mamba: Dell’s Tyler Cox and Shirish Gupta discuss with Jon Krohn why state space models can process information so efficiently, and how Dell’s AI factory helps enterprises manage custom AI workloads. Hear the latest on the Dell Pro AI Studio and Dell’s partnerships with IBM and Hugging Face in this episode. 

This episode is brought to you by the Trainium2, the latest AI chip from AWS and by Gurobi.

Additional materials: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠www.superdatascience.com/939⁠

Interested in sponsoring a SuperDataScience Podcast episode? Email natalie@superdatascience.com for sponsorship information.

In this episode you will learn:

• (02:58) Dell Pro AI Studio news
• (23:17) How Dell manages interoperability
• (28:08) About the Dell/IBM granite models
• (47:38) How to troubleshoot AI tools
• (52:36) How Dell performs against benchmarks