# Google DeepMind — EmbeddingGemma 2: an open, lightweight multimodal embedding model

- Company: Google DeepMind (deepmind.google)
- Announced: 2026-10-06T19:57:04+00:00
- Category: not stated
- Coverage: 8 outlets
- Announcement: yes
- Group: covered
- Source: https://deepmind.google/blog/embeddinggemma-2-an-open-lightweight-multimodal-embedding-model/
- Record: https://forck.live/items/16263-embeddinggemma-2-an-open-lightweight-multimodal-embedding-model
- Subject: Gemini

EmbeddingGemma 2 is the most capable model for on-device multimodal embeddings, natively mapping combinations of text, images, audio, and video into a unified embedding space. Sahil Dua Research Engineer, Google DeepMind Henrique Schechter Vera Research Engineer, Google DeepMind We introduced EmbeddingGemma last year to provide a lightweight option for high-quality text embeddings, to help your apps organize, search, and connect information directly on consumer hardware. The developer community’s response blew past our expectations. With more than 20 million downloads, builders have used it to power smarter on-device search tools and privacy-first retrieval augmented generation (RAG) pipelines. Today, we’re launching EmbeddingGemma 2 , expanding beyond text to unify code, images, video, and audio in a shared embedding space. Built on the Gemma 4 architecture and released under a commercially permissive Apache 2.0 license, EmbeddingGemma 2 has 740 million parameters, making it optimal for on-device inference. …

## Around this story

Outlets this record can name and link:

- The Verge — Google launches the next version of its on-device AI model.: https://theverge.com/tech/1005886/google-launches-the-next-version-of-its-on-device-ai-model
- The New Stack — Your phone's vector index might be bigger than the AI model running it: https://thenewstack.io/google-embeddinggemma-multimodal-search
- Unite — DeepMind Debuts EmbeddingGemma 2, Mapping Five Modalities Into One Space: https://unite.ai/deepmind-debuts-embeddinggemma-2-mapping-five-modalities-into-one-space
- Blog — EmbeddingGemma 2: an open, lightweight multimodal embedding model: https://blog.google/innovation-and-ai/technology/developers-tools/embeddinggemma-2
- The Next Web — Google launches EmbeddingGemma 2, an open multimodal embedding model for devices: https://thenextweb.com/news/embeddinggemma-2-on-device-europe
- Google Developers Blog — Bring multimodal semantic search to the edge with EmbeddingGemma 2: https://developers.googleblog.com/en/google-ai-edge-with-embeddinggemma-2
- Unsloth — EmbeddingGemma 2 - Run Locally | Unsloth Documentation: https://unsloth.ai/docs/models/embeddinggemma-2

## Follow-ups

- The Decoder (reported, 6 Oct): Google claims EmbeddingGemma 2 outperforms rival embedding models twice its size — https://forck.live/items/16260-google-claims-embeddinggemma-2-outperforms-rival-embedding-models-twice-its-size

---

Record: https://forck.live/items/16263-embeddinggemma-2-an-open-lightweight-multimodal-embedding-model
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md