Google DeepMind
6 October 2026
EmbeddingGemma 2
Context: —Input: — / 1MOutput: — / 1M
What's new
- Creates embeddings for text, code, images, video, and audio in a shared 768-dimensional space.
- Designed to run on-device; the text component uses about 191 MB of RAM on a Pixel 11 Pro, while the full multimodal version uses about 567 MB.