Here’s everything that happened this week 🚀: — @GoogleMaps released 2 new features, Ask Maps to handle your most complex questions about places and trips and Immersive Navigation for intuitive routes, all with some help from the latest Gemini models — New Gemini features rolled out to @GoogleWorkspace, making @GoogleDocs, Sheets, Slides, and @GoogleDrive more helpful — In collaboration with Imperial College London and the UK’s NHS, we published breast cancer research that demonstrates AI’s potential to detect 25% of interval cancers previously missed by conventional methods — We introduced Gemini Embedding 2 (in preview), our first natively multimodal embedding model, which enables semantic understanding across text, images, videos, audio, and documents inputs all in a single model — We also launched project spend caps for the Gemini API in @GoogleAIStudio, enabling you to set a dollar amount for maximum spend at https://t.co/NApz8LVHll — Gemini in @GoogleChrome began rolling out on desktop to signed-in users (18+) in India, New Zealand, and Canada, with expansions to mobile and more regions and languages coming throughout the year
Google Launches Gemini Embedding 2, First Natively Multimodal Embedding Model
Google· Updated
Gemini Embedding 2, now in preview via the Gemini API, is Google's first natively multimodal embedding model — enabling semantic understanding across text, images, videos, audio, and documents in a unified representation space.
gemini-embedding-2-preview), now in preview, is Google’s first natively multimodal embedding model — accepting text, images, videos, audio, and documents and producing embeddings that capture semantic meaning across all input types in a single model.Most production retrieval and search systems handle modalities separately — text embeddings in one system, image embeddings in another — requiring stitched-together pipelines for cross-modal queries. Gemini Embedding 2 collapses that into a unified representation space, making cross-modal semantic search achievable in a single model.
Access it now through the Gemini API in preview — try it against a mixed-content corpus to test cross-modal retrieval across your existing content types.
Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

