This Day in AI Podcast cover image

EP44: The Finale: Google Gemini, SimTheory, Is Ilya OK? Predictions for 2024

This Day in AI Podcast

00:00

Training a Model on Multimodal Tokens

This chapter discusses the training process of a model on over 1.5 trillion multimodal tokens from web documents, books, code, images, audio, and videos. The speakers express their interest in the video capability and audio integration of the model. They suggest that if the creators were more fearless and released the model openly, it could have showcased Google's leadership in the AI space.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app