embeddinggemma-2

What is embeddinggemma-2

EmbeddingGemma-2 is a text embedding model series released by Google, focused on generating high-quality text vector representations for tasks such as search, retrieval and similarity computation.

Key features

• Maps text (including code), images, video and audio inputs into a unified 768-dimensional vector space
• Supports 100+ languages and improves performance on code-related tasks
• Provides flexible modality loading, allowing selective use of text, vision or audio encoders
• Natively supports Matryoshka Representation Learning, allowing truncation from 128d to 768d embeddings
• Supports an 8K token context window, capable of handling several minutes of audio or video

Who it is for

Suitable for developers needing to perform text, image, video and audio embedding on consumer-grade hardware (such as phones and laptops) for search, RAG, classification and clustering tasks.

How to get started

1. Visit https://huggingface.co/google/embeddinggemma-2
2. Load the model using the Hugging Face Transformers library
3. Selectively load text, vision or audio encoders according to requirements

Compiled from public descriptions on the tool's official website / README / launch page. Pricing and user numbers have not been verified; please check the official website. Some sections were drafted with AI assistance. Written in Chinese and English; other languages are machine-translated.