Google DeepMind releases EmbeddingGemma 2
AIMade by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made
Google DeepMind released EmbeddingGemma 2 — a 740 million-parameter open model (Gemma 4 architecture, Apache 2.0) that maps text, code, images, audio and video into one embedding space
Why it matters
Company-reported MTEB Code rose from 68.76 to 78.68; context is 8K (4 times its predecessor), and MRL lets the 768 dimensions shrink to 128, cutting storage by up to 6 times. Quantized, it runs on a Pixel 11 Pro in about 191MB of RAM for text only and about 567MB for the full multimodal model, bringing on-device multimodal search to open weights
Source — Primary sources first; secondary reports are labelled “Press”
Officialdeepmind.googlehttps://deepmind.google/blog/embeddinggemma-2-an-open-lightweight-multimodal-embedding-model/Added in Week 42, 2026 · See that week
On the map — ⑰ Embeddings line
- Previous stopvoyage-context-4
- This record · stopEmbeddingGemma 2
- Next stopNo later stop yet
In the model family tree — Gemma family · 7 versions
- Earlier versionDiffusionGemma
- This recordEmbeddingGemma 2
- Later versionThe latest version in this branch