Office Hours with Baris Gultekin: The Era of Enterprise AI is Here
Office Hours with Tomasz Tunguz
00:00
Innovations in Model Architecture for Cost-Effective AI
This chapter explores a new model architecture designed to cut down training and inference costs in large language models. It emphasizes the combination of dense models and mixture of experts to enhance efficiency and discusses the strategic use of large models in early project stages, optimizing later with smaller models.
Transcript
Play full episode
