---
format: "aidr-story-markdown/v1"
id: "66213f22c321bd47b9dab44448ee4cf8f6d510ea5854421e125147c98fa5bb3b"
canonical_url: "https://aidr.today/66213f22?lang=en"
title: "Google claims EmbeddingGemma 2 outperforms rival embedding models twice its size"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-06T19:47:27.000Z"
category: "Data"
topics: ["google","open-source","data-engineering"]
source_urls: ["https://the-decoder.com/google-claims-embeddinggemma-2-outperforms-rival-embedding-models-twice-its-size/","https://deepmind.google/blog/embeddinggemma-2-an-open-lightweight-multimodal-embedding-model/","https://blog.google/innovation-and-ai/technology/developers-tools/embeddinggemma-2/","https://simonwillison.net/2026/Oct/6/hn-49983751/","https://news.ycombinator.com/item?id=49980487"]
summary: "Google has released EmbeddingGemma 2, an open model with 740 million parameters that converts text, images, video, audio, and code into vectors. The model runs on-device, requires about 191 MB of RAM, and reportedly outperforms some competing models twice its size. When paired with a small open model like Gemma 4, it can run offline RAG apps without sending data to external servers."
---

# Google claims EmbeddingGemma 2 outperforms rival embedding models twice its size

> [Open the canonical story](<https://aidr.today/66213f22?lang=en>)

**Published:** 2026-10-06T19:47:27.000Z
**Category:** Data
**Topics:** google, open\-source, data\-engineering

## Summary

Google has released EmbeddingGemma 2, an open model with 740 million parameters that converts text, images, video, audio, and code into vectors\. The model runs on\-device, requires about 191 MB of RAM, and reportedly outperforms some competing models twice its size\. When paired with a small open model like Gemma 4, it can run offline RAG apps without sending data to external servers\.

## Sources

- [Story source](<https://the-decoder.com/google-claims-embeddinggemma-2-outperforms-rival-embedding-models-twice-its-size/>)
- [Story source](<https://deepmind.google/blog/embeddinggemma-2-an-open-lightweight-multimodal-embedding-model/>)
- [Story source](<https://blog.google/innovation-and-ai/technology/developers-tools/embeddinggemma-2/>)
- [Story source](<https://simonwillison.net/2026/Oct/6/hn-49983751/>)
- [Discussion](<https://news.ycombinator.com/item?id=49980487>)