Google launched Gemini Embedding 2, a multimodal artificial intelligence model designed to integrate diverse data types into a single unified space. The model maps text, images, video, audio, and documents simultaneously. It captures semantic intent across more than 100 languages.
The architecture simplifies complex data pipelines for Retrieval-Augmented Generation (RAG) and semantic search. It also enhances sentiment analysis and data clustering tasks. Google reports that the model establishes a new performance standard, outperforming leading competitors in text, image, and video processing.
Alphabet shares recorded a fractional increase in midday trading following the announcement. This release marks a strategic move to maintain competitiveness in the evolving AI sector.