
Optimizing for efficiency with IBM’s Granite
Practical AI
00:00
Advancements in AI: The Granite Model Revolution
This chapter explores the evolution of AI through the release of Granite 3.2, featuring enhanced reasoning capabilities and the innovative 'inference time compute' feature. It discusses the shift towards smaller, efficient models while still maintaining larger ones for complex tasks, and highlights the introduction of safety mechanisms within IBM's Granite framework. The conversation also covers the transition to edge computing, emphasizing the importance of optimizing models and prompt design for improved performance.
Play episode from 18:09
Transcript


