🚨 Google just dropped EmbeddingGemma 2, a 740M-parameter multimodal embedding model built for on-device AI It can connect text, images, audio and video in the same space, meaning you can search an image library with text, find a video using a voice memo, or search long audio