Google unifies text, image, video, and audio in a single vector space with Gemini Embedding 2
The Decoder
•
Generative AI
Google's first native multimodal embedding model brings text, images, video, audio, and documents into one vector space, cutting out the need for separate models in AI pipelines.