Report
Google's EmbeddingGemma 2 brings multimodal search to one on-device model
The New Stack says developers can skip captioning and transcription when searching text, code, images, video and audio.
TLDR
The New Stack reports that Google's EmbeddingGemma 2 brings text, code, image, video and audio search into one on-device model, letting developers skip captioning and transcription. It also notes that a phone's vector index might be bigger than the AI model running it.
Combined views
231
1 Source, first seen ago