Grilled Cheese

ExploreLog inSign up
Terms of UsePrivacy PolicyCommunity StandardsHelpGet the app

Grilled Cheese is a product of Village Compute

Version devBuilt at: 2026-10-10 01:38:52 EDT

Explore

PostsPeople
LatestRanked
@eamagician.bsky.socialOct 7, 2026, 10:35 PM

RAGの埋め込みを外部APIに投げるの、そろそろ恥ずかしくなってきた。GoogleのEmbeddingGemma 2は740MでApache 2.0。テキスト・画像・音声・動画・コードを768次元に落とす。RAM約191MB。小型モデルと組めばオフラインRAG。データを外に出さない安心感は、精度の数%より効く場面がある。個人開発の検索は一旦これで固定かな。#EmbeddingGemma

@pxology.bsky.socialOct 7, 2026, 12:07 PM

Google, 740 milyon parametreli açık model EmbeddingGemma 2’yi yayımladı.

Metin, kod, görsel, video ve sesi ortak bir gömme uzayında işliyor; modüler yapısıyla cihaz üzerinde çalışmaya odaklanıyor.

#EmbeddingGemma #Gemma #GoogleAI #OpenSourceAI #OnDeviceAI

Detaylar:
blog.google/innovation-a...

@newzblock.bsky.socialOct 7, 2026, 11:08 AM

🔦 UNDER-REPORTED — barely anyone else is covering this:

Google launches EmbeddingGemma 2, a li

#EmbeddingGemma #GoogleAI #MachineLearning #MobileAI #TechNews #AI #Innovation #Breaking

Read more: https://newzblock.com/story/finetazo-la-reapropiacion-comunitaria-del-malecon-de-santo-domingo-9414f9

🔦 UNDER-REPORTED — barely anyone else is covering this: Google launches EmbeddingGemma 2, a lightweight multimodal model for searching photos, videos, and audio offline on mobile Google DeepMind has introduced EmbeddingGemma 2, a new lightweight, open-weights multimodal model designed for develop🔦 UNDER-REPORTED — barely anyone else is covering this: Google launches EmbeddingGemma 2, a lightweight multimodal model for searching photos, videos, and audio offline on mobile Google DeepMind has introduced EmbeddingGemma 2, a new lightweight, open-weights multimodal model designed for develop🔦 UNDER-REPORTED — barely anyone else is covering this: Google launches EmbeddingGemma 2, a lightweight multimodal model for searching photos, videos, and audio offline on mobile Google DeepMind has introduced EmbeddingGemma 2, a new lightweight, open-weights multimodal model designed for develop🔦 UNDER-REPORTED — barely anyone else is covering this: Google launches EmbeddingGemma 2, a lightweight multimodal model for searching photos, videos, and audio offline on mobile Google DeepMind has introduced EmbeddingGemma 2, a new lightweight, open-weights multimodal model designed for develop
@eamagician.bsky.socialOct 6, 2026, 10:42 PM

RAGの品質は、ぶっちゃけembeddingで8割決まる。GoogleのEmbeddingGemma 2は740Mで、量子化すれば191MB。テキスト・画像・音声・動画・コードを全部同じ768次元に落とせる。小さいモデルと組めば外部に何も送らずローカルでRAGが回る。華はないけど、こういう地味なのが結局一番長く使うやつ。#EmbeddingGemma

@logeshwaranorg.bsky.socialOct 6, 2026, 4:52 PM

#Google just released #EmbeddingGemma 2: one open 740M model for text, code, images, video and audio in one search space. 378 MB text-only in Ollama, Apache 2.0, runs on a phone.

www.logeshwaran.org/2026/10/embe... #AI #Opensource #LocalLLM