I love EmbeddingGemma 2 and would die for EmbeddingGemma 2. I don't need sophisticated organizational systems to keep all the established lore in my webnovel straight anymore, I just need EmbeddingGemma and a robot to query the embeddings!!
Check out this practical developer guide for EmbeddingGemma 2 covering multimodal retrieval across text, code, images, audio, and video, modular encoders (270M to 740M), and Matryoshka truncation (768d down to 128d):
https://developers.googleblog.com/embeddinggemma-2-the-developer-guide/
https://developers.googleblog.com/embeddinggemma-2-the-developer-guide/
October 12, 2026 at 2:54 PM
I love EmbeddingGemma 2 and would die for EmbeddingGemma 2. I don't need sophisticated organizational systems to keep all the established lore in my webnovel straight anymore, I just need EmbeddingGemma and a robot to query the embeddings!!
Check out this practical developer guide for EmbeddingGemma 2 covering multimodal retrieval across text, code, images, audio, and video, modular encoders (270M to 740M), and Matryoshka truncation (768d down to 128d):
https://developers.googleblog.com/embeddinggemma-2-the-developer-guide/
https://developers.googleblog.com/embeddinggemma-2-the-developer-guide/
October 12, 2026 at 1:18 PM
Check out this practical developer guide for EmbeddingGemma 2 covering multimodal retrieval across text, code, images, audio, and video, modular encoders (270M to 740M), and Matryoshka truncation (768d down to 128d):
https://developers.googleblog.com/embeddinggemma-2-the-developer-guide/
https://developers.googleblog.com/embeddinggemma-2-the-developer-guide/
Oh wow, embeddinggemma is a much better model for my needs.
October 14, 2026 at 7:17 PM
Oh wow, embeddinggemma is a much better model for my needs.
again, the power of the community always makes me happy.
we launched EmbeddingGemma 2 last week, and people already built a local search system that is better than the integrated mac one!!
www.reddit.com/r/LocalLLaMA...
we launched EmbeddingGemma 2 last week, and people already built a local search system that is better than the integrated mac one!!
www.reddit.com/r/LocalLLaMA...
From the LocalLLaMA community on Reddit: Open-source Mac app that runs EmbeddingGemma 2 locally to search your files by what’s in them
Explore this post and more from the LocalLLaMA community
www.reddit.com
October 11, 2026 at 11:04 PM
again, the power of the community always makes me happy.
we launched EmbeddingGemma 2 last week, and people already built a local search system that is better than the integrated mac one!!
www.reddit.com/r/LocalLLaMA...
we launched EmbeddingGemma 2 last week, and people already built a local search system that is better than the integrated mac one!!
www.reddit.com/r/LocalLLaMA...
seeing more and more cool machine learning tools that aren’t LLMs lately; case in point: EmbeddingGemma 2, a small (<1gb RAM) model that takes text, images, and audio and returns a high dimensional vector for comparing similarity between disparate media
blog.google/innovation-a...
blog.google/innovation-a...
EmbeddingGemma 2: an open, lightweight multimodal embedding model
Introducing EmbeddingGemma 2, an open multimodal embedding model optimized for privacy-first use cases
blog.google
October 7, 2026 at 1:55 AM
seeing more and more cool machine learning tools that aren’t LLMs lately; case in point: EmbeddingGemma 2, a small (<1gb RAM) model that takes text, images, and audio and returns a high dimensional vector for comparing similarity between disparate media
blog.google/innovation-a...
blog.google/innovation-a...
DigUp lets users find Mac files by describing them, not remembering their names #Technology #SoftwareandApps #MobileApps #digup #macapp #embeddinggemma #naturalanguage #opensource
https://puretech.news/read?id=235921219939991552
https://puretech.news/read?id=235921219939991552
puretech.news
October 14, 2026 at 5:38 AM
DigUp lets users find Mac files by describing them, not remembering their names #Technology #SoftwareandApps #MobileApps #digup #macapp #embeddinggemma #naturalanguage #opensource
https://puretech.news/read?id=235921219939991552
https://puretech.news/read?id=235921219939991552
It simply mystifies me how all these labs release Matryoshka versions and fail to offer quantized vectors: Remex, based on Google’s own TurboQuant offers FAR superior byte-performance: 95% at 2bit — yet another case of a 16x reduction. See below for results:
muninn.austegard.com/blog/sixteen...
muninn.austegard.com/blog/sixteen...
October 11, 2026 at 12:53 AM
It simply mystifies me how all these labs release Matryoshka versions and fail to offer quantized vectors: Remex, based on Google’s own TurboQuant offers FAR superior byte-performance: 95% at 2bit — yet another case of a 16x reduction. See below for results:
muninn.austegard.com/blog/sixteen...
muninn.austegard.com/blog/sixteen...
Fun interactive demo for EmbeddingGemma 2: sketch a target object in the browser and the model scores your drawing in real time against distractors until you hit 90% cosine similarity in the shared text-image vector space. https://huggingface.co/spaces/google/embedding-draw-challenge
October 14, 2026 at 9:32 AM
Fun interactive demo for EmbeddingGemma 2: sketch a target object in the browser and the model scores your drawing in real time against distractors until you hit 90% cosine similarity in the shared text-image vector space. https://huggingface.co/spaces/google/embedding-draw-challenge
I’ve never had any kind of search feature on my website but EmbeddingGemma has me thinking about implementing a possibly truly blursed semantic search for my blog 🤔
October 14, 2026 at 6:05 PM
I’ve never had any kind of search feature on my website but EmbeddingGemma has me thinking about implementing a possibly truly blursed semantic search for my blog 🤔
EmbeddingGemma 2 Integration by @elbruno.com
https://elbruno.com/2026/10/08/embeddinggemma-2-integration/?utm_source=bluesky&utm_medium=social&utm_campaign=newsletter
https://elbruno.com/2026/10/08/embeddinggemma-2-integration/?utm_source=bluesky&utm_medium=social&utm_campaign=newsletter
EmbeddingGemma 2 Integration by @elbruno.com
Hi! EmbeddingGemma 2 Integration ElBruno.LocalEmbeddings.EmbeddingGemma provides local text-only embeddings from Google’s EmbeddingGemma 2 model through IEmbeddingGenerator<string, Embeddi…
elbruno.com
October 12, 2026 at 6:35 PM
EmbeddingGemma 2 Integration by @elbruno.com
https://elbruno.com/2026/10/08/embeddinggemma-2-integration/?utm_source=bluesky&utm_medium=social&utm_campaign=newsletter
https://elbruno.com/2026/10/08/embeddinggemma-2-integration/?utm_source=bluesky&utm_medium=social&utm_campaign=newsletter
Thanks Gus. Serious question: is there any reason the team did not release a a vector quantizer? In my empirical testing, quantizing the vectors (w a Google TurboQuant derivative!) all the way down to 2bit VASTLY outperforms a Matryoshka version: muninn.austegard.com/blog/sixteen...
A 16× Smaller Index for EmbeddingGemma 2
EmbeddingGemma 2’s one advertised way to shrink its vectors, keeping 128 of 768 dimensions, loses about 0.06 nDCG@10. Storing all 768 at 2 bits each is 16× smaller with no measurable loss on SciFact o...
muninn.austegard.com
October 13, 2026 at 2:10 AM
Thanks Gus. Serious question: is there any reason the team did not release a a vector quantizer? In my empirical testing, quantizing the vectors (w a Google TurboQuant derivative!) all the way down to 2bit VASTLY outperforms a Matryoshka version: muninn.austegard.com/blog/sixteen...
EmbeddingGemma 2: an open, lightweight multimodal embedding model #machinelearning #ai
EmbeddingGemma 2: an open, lightweight multimodal embedding model
Groups
Conversations
All groups and messages
Sign in
EmbeddingGemma 2: an open, lightweight multimodal embedding model
0 views
Eyal Estrin
unread,
11:25 PM (9 minutes ago)
to
https://blog.google/innovation-and-ai/technology/developers-tools/embeddinggemma-2/
https://developers.googleblog.com/google-ai-edge-with-embeddinggemma-2/
https://developers.googleblog.com/embeddinggemma-2-the-developer-guide/
https://9to5google.com/2026/10/06/google-ai-edge-foresight/
Eyal Estrin
Author | Cloud Architect | AWS • Azure • GCP Insights
Social: @eyalestrin
Connect: https://linktr.ee/eyalestrin Blog: https://security-24-7.com
Reply all
Reply to author
Forward
dlvr.it
October 9, 2026 at 3:34 AM
EmbeddingGemma 2: an open, lightweight multimodal embedding model #machinelearning #ai
EmbeddingGemma 2, a tiny multimodal mobile ready embedding model
EmbeddingGemma 2: an open, lightweight multimodal embedding model
Introducing EmbeddingGemma 2, an open multimodal embedding model optimized for privacy-first use cases
blog.google
October 7, 2026 at 6:52 AM
EmbeddingGemma 2, a tiny multimodal mobile ready embedding model
GoogleのEmbeddingGemma 2を音声認識モデルに変換した「Embedding Gemma Transcribe」、オフライン環境のMacなら15分の音声をわずか約3秒で爆速文字起こし可能
https://gigazine.net/news/20261014-embetting-gemma-transcribe/
https://gigazine.net/news/20261014-embetting-gemma-transcribe/
GoogleのEmbeddingGemma 2を音声認識モデルに変換した「Embedding Gemma Transcribe」、オフライン環境のMacなら15分の音声をわずか約3秒で爆速文字起こし可能
音声をテキストに変換する文字起こしには一般的に音声認識専用のAIモデルが使われます。しかし、文章の意味を数値で表現する埋め込みモデルのEmbeddingGemma 2を利用して音声を文字起こしするという一風変わったアプローチを採用した「Embedding Gemma Transcribe」が登場しました。Embedding Gemma TranscribeはApple Silicon搭載Macで動作し、インターネット接続なしで音声を処理できるとのことなので、実際に使ってみました。
gigazine.net
October 14, 2026 at 1:01 PM
GoogleのEmbeddingGemma 2を音声認識モデルに変換した「Embedding Gemma Transcribe」、オフライン環境のMacなら15分の音声をわずか約3秒で爆速文字起こし可能
https://gigazine.net/news/20261014-embetting-gemma-transcribe/
https://gigazine.net/news/20261014-embetting-gemma-transcribe/
We turned EmbeddingGemma 2 into a speech-to-text model
https://laxima.tech/signal/we-turned-embeddinggemma-2-into-a-speech-to-text-model-hn-50069971
https://laxima.tech/signal/we-turned-embeddinggemma-2-into-a-speech-to-text-model-hn-50069971
October 13, 2026 at 6:17 PM
We turned EmbeddingGemma 2 into a speech-to-text model
https://laxima.tech/signal/we-turned-embeddinggemma-2-into-a-speech-to-text-model-hn-50069971
https://laxima.tech/signal/we-turned-embeddinggemma-2-into-a-speech-to-text-model-hn-50069971
ニュース「ファイルの内容から探せるMac向けOSSアプリ「DigUp」、EmbeddingGemma 2をローカル実行」公開 gihyo.jp/article/2026...
ファイルの内容から探せるMac向けOSSアプリ「DigUp」、EmbeddingGemma 2をローカル実行 | gihyo.jp
開発者のAbdur Rahim氏は2026年10月9日、Mac内のファイルを内容から検索するアプリDigUpをXで紹介した。
gihyo.jp
October 14, 2026 at 11:41 AM
ニュース「ファイルの内容から探せるMac向けOSSアプリ「DigUp」、EmbeddingGemma 2をローカル実行」公開 gihyo.jp/article/2026...
This is exciting. Basically multimodal RAG over pictures, arbitrary text, web, voice that runs on your phone and does not share anything with the cloud. A large number of desktop and mobile privacy preserving apps can be built on this.
blog.google/innovation-a...
#AI
blog.google/innovation-a...
#AI
EmbeddingGemma 2: an open, lightweight multimodal embedding model
Introducing EmbeddingGemma 2, an open multimodal embedding model optimized for privacy-first use cases
blog.google
October 14, 2026 at 6:05 AM
This is exciting. Basically multimodal RAG over pictures, arbitrary text, web, voice that runs on your phone and does not share anything with the cloud. A large number of desktop and mobile privacy preserving apps can be built on this.
blog.google/innovation-a...
#AI
blog.google/innovation-a...
#AI
🤗 Climbing on Hugging Face right now: google/embeddinggemma-2
⬇ 115.8k downloads / 30 days · ♡ 2k · feature extraction · Apache-2.0
Open-weight models ranked by real-world usage, every day:
https://olud.ai/#leaderboard
#OpenSource #AI #HuggingFace
⬇ 115.8k downloads / 30 days · ♡ 2k · feature extraction · Apache-2.0
Open-weight models ranked by real-world usage, every day:
https://olud.ai/#leaderboard
#OpenSource #AI #HuggingFace
October 13, 2026 at 11:02 PM
🤗 Climbing on Hugging Face right now: google/embeddinggemma-2
⬇ 115.8k downloads / 30 days · ♡ 2k · feature extraction · Apache-2.0
Open-weight models ranked by real-world usage, every day:
https://olud.ai/#leaderboard
#OpenSource #AI #HuggingFace
⬇ 115.8k downloads / 30 days · ♡ 2k · feature extraction · Apache-2.0
Open-weight models ranked by real-world usage, every day:
https://olud.ai/#leaderboard
#OpenSource #AI #HuggingFace
Google DeepMind、「EmbeddingGemma 2」を発表 ～テキストから動画まで扱えるオープン埋め込みモデル – 窓の杜
https://www.yayafa.com/2908514/
「EmbeddingGemma 2」…
https://www.yayafa.com/2908514/
「EmbeddingGemma 2」…
Google DeepMind、「EmbeddingGemma 2」を発表 ～テキストから動画まで扱えるオープン埋め込みモデル - 窓の杜 - YAYAFA
「EmbeddingGemma 2」
www.yayafa.com
October 12, 2026 at 2:40 AM
Google DeepMind、「EmbeddingGemma 2」を発表 ～テキストから動画まで扱えるオープン埋め込みモデル – 窓の杜
https://www.yayafa.com/2908514/
「EmbeddingGemma 2」…
https://www.yayafa.com/2908514/
「EmbeddingGemma 2」…
What if Spotlight actually understood the meaning inside your files?
DigUp is an open-source macOS menu bar app powered by EmbeddingGemma 2 that brings 100% local, multimodal semantic search to your Mac.
Instead of searching for file names, you can use natural language to:
DigUp is an open-source macOS menu bar app powered by EmbeddingGemma 2 that brings 100% local, multimodal semantic search to your Mac.
Instead of searching for file names, you can use natural language to:
October 13, 2026 at 4:51 PM
What if Spotlight actually understood the meaning inside your files?
DigUp is an open-source macOS menu bar app powered by EmbeddingGemma 2 that brings 100% local, multimodal semantic search to your Mac.
Instead of searching for file names, you can use natural language to:
DigUp is an open-source macOS menu bar app powered by EmbeddingGemma 2 that brings 100% local, multimodal semantic search to your Mac.
Instead of searching for file names, you can use natural language to:
Friend of mine just introduced me to this and I will test and let you all know how well it works ig.
GitHub - ARahim3/DigUp: Search your Mac by what's in your files: photos, PDFs, recordings, videos and code. On-device, built on Google DeepMind's EmbeddingGemma 2.
Search your Mac by what's in your files: photos, PDFs, recordings, videos and code. On-device, built on Google DeepMind's EmbeddingGemma 2. - ARahim3/DigUp
github.com
October 14, 2026 at 5:42 PM
Friend of mine just introduced me to this and I will test and let you all know how well it works ig.
Google released EmbeddingGemma 2 - a new multimodel embeddings open weights model. Looks supper cool 👇🏼
www.youtube.com/watch?v=anPs...
www.youtube.com/watch?v=anPs...
Introducing EmbeddingGemma 2: An open model for natively multimodal embeddings
YouTube video by Google for Developers
www.youtube.com
October 8, 2026 at 5:45 AM
Google released EmbeddingGemma 2 - a new multimodel embeddings open weights model. Looks supper cool 👇🏼
www.youtube.com/watch?v=anPs...
www.youtube.com/watch?v=anPs...
The model: huggingface.co/google/embed...
Give it a like if you haven't already, or even better: give it a try, it's natively integrated in Sentence Transformers for any modality you want.
Give it a like if you haven't already, or even better: give it a try, it's natively integrated in Sentence Transformers for any modality you want.
google/embeddinggemma-2 · Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
October 13, 2026 at 11:48 AM
The model: huggingface.co/google/embed...
Give it a like if you haven't already, or even better: give it a try, it's natively integrated in Sentence Transformers for any modality you want.
Give it a like if you haven't already, or even better: give it a try, it's natively integrated in Sentence Transformers for any modality you want.
Google、画像や動画にも対応する「EmbeddingGemma 2」を発表 – AI Watch
https://www.yayafa.com/2908960/
オープンマルチモーダルモデル「Emb…
https://www.yayafa.com/2908960/
オープンマルチモーダルモデル「Emb…
Google、画像や動画にも対応する「EmbeddingGemma 2」を発表 - AI Watch - YAYAFA
オープンマルチモーダルモデル「EmbeddingGemma 2」
www.yayafa.com
October 12, 2026 at 3:00 PM
Google、画像や動画にも対応する「EmbeddingGemma 2」を発表 – AI Watch
https://www.yayafa.com/2908960/
オープンマルチモーダルモデル「Emb…
https://www.yayafa.com/2908960/
オープンマルチモーダルモデル「Emb…