Google DeepMind details EmbeddingGemma 2: unified embeddings for code, image, audio, video
GoogleDeepMind · x · 2026-10-07
Google DeepMind officially introduced EmbeddingGemma 2, its first open, natively multimodal embedding model built for on-device use.
- Goes beyond text: unifies code, images, audio, and video in a shared embedding space
- Lightweight and modular, aimed at efficient on-device deployment
Same announcement as Pichai's post; details live in the official thread.
More from Models
- vLLM adds day-0 support for Google's multimodal EmbeddingGemma 2 — AccBalanced · 2026-10-07
- RedNote Has a Serious, Low-Key AI Lab, Argues Analyst Weighing Dots3 Results — teortaxesTex · 2026-10-07
- OpenAI Cuts API Rate Limit Tiers From Five to Three, Grow Tier Now at $500 — OpenAIDevs · 2026-10-07
- Users report GPT-6.1 Sol is slower and worse in practice, suspect sandbagging — rickasaurus · 2026-10-07
- 61–90% of AI-synthesized medical conclusions contain factual errors, SciConBench finds — manoelribeiro · 2026-10-07
- Cohere Labs Debuts Tiny Aya, a Small-Model Family Covering 70+ Languages — Cohere_Labs · 2026-10-07