imagebind
Meta project for shared embedding space across modalities
Episodes
-
#4774: When a Token Isn't a Word AnymoreImages, audio, and code all use "tokens" — but they're completely different things. Here's how tokenization actually works across modalities. -
#1792: Google's Native Multimodal Embedding Kills the Fusion LayerGoogle’s new embedding model maps text, images, audio, and video into a single vector space—cutting latency by 70%.