EmbeddingGemma 2: an open, lightweight multimodal embedding model

What changed

EmbeddingGemma 2: an open, lightweight multimodal embedding model We introduced EmbeddingGemma last year to provide a lightweight option for high-quality text embeddings, to help your apps organize, search, and connect information directly on consumer hardware. Today, we’re launching EmbeddingGemma 2, expanding beyond text to unify code, images, video, and audio in a shared embedding space. Built on the Gemma 4 architecture and released under a commercially permissive Apache 2.0 license, EmbeddingGemma 2 has 740 million parameters, making it optimal for on-device inference.

Why it matters

A concrete addition to Practical AI: Tools, Models & Frameworks: it changes what's available to builders today rather than being general commentary.

How it compares

Related prior coverage to compare against:

• Multimodal Embedding & Reranker Models with Sentence Transformers
• New embedding models and API updates
• Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers

• Multimodal Embedding & Reranker Models with Sentence Transformers similarity 0.31
• New embedding models and API updates similarity 0.27
• Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers similarity 0.26