Introducing EmbeddingGemma 2: An open model for natively multimodal embeddings @GoogleDevelopers
Introducing EmbeddingGemma 2: An open model for natively multimodal embeddings  @GoogleDevelopers
Uploaded October 2026 | Updated November 2026, 3 days ago
Sebastian Russo, Ivan Llanos, and Sahil Dua from Google DeepMind introduce EmbeddingGemma 2, a lightweight open model that maps text, images, video, and audio into a single unified embedding space.

At just 740 million parameters, EmbeddingGemma 2 is engineered for on-device use cases. Watch this video to discover:

*EmbeddingGemma 2’s modular architecture with optional audio and vision encoders
*Instant media search and video moment retrieval running fully offline on device
*Retrieval-augmented generation (RAG) pipelines where sensitive data never leaves the hardware
*Fine-tuning for domain-specific use cases


πŸ”— Resources:
Learn about the model β†’ https://goo.gle/4hE3Lb9
Read the developer guide β†’ https://goo.gle/4hFnyqO
Explore on-device embeddings with Google AI Edge β†’ https://goo.gle/3VSXqS1
Download model weights from HuggingFace β†’ https://goo.gle/4i45qbp
Get started with the Gemma Cookbook β†’ https://goo.gle/3TCsqVO

Have questions? Drop them in the comments.



Subscribe to Google for Developers β†’ https://goo.gle/developers



Speakers: Sebastian Russo, Ivan Llanos, Sahil Dua
Products Mentioned: Gemma, Google AI Edge
Introducing EmbeddingGemma 2: An open model for natively multimodal embeddingsDeveloper Keynote (Google I/O 26)Announcing Gemma 4, Google I/O β€˜26 Updates, and more! - Google Developer News April 2026A Git challenge: Which version ends up in the repo?Defining the agentic AI eraGemma Playground: Parallel Agents in Action3 takeaways on Software Ecologyℒ️Voice Agent observability with LangSmith 🌟Bring the power of on-device AI to life with Google AI Edge and GemmaBuilding the ultimate morning dashboard with Gemini 3.6 Flash and Nano BananaAI tools for human creativityπŸͺ΄ Building an autonomous plant rover with Gemini Live
Google for Developers |

Introducing EmbeddingGemma 2: An open model for natively multimodal embeddings

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER