Unlocking innovative generative AI use cases with text and multimodal embeddings

Embeddings power some of the most commercially widespread applications of AI, such as semantic search and recommender systems. In this session, we explain why embeddings are a critical component of building generative AI applications. You learn about common architectural patterns for using Embeddings and how to use them in conjunction with vector databases. We also deep dive into multimodal embeddings and how you can use them to unlock a whole array of new applications such as recommendations and content classification.

Speakers: Anand Iyer, Eric Poff

Watch more:

#GoogleCloudNext

[looppp]

I think I found a bug for the video embedding part of your Multimodal Embedding Model.

I keep getting a "Deadline" error when embedding a video. No issues with image or contextual_text.