What is EmbeddingGemma 2?

EmbeddingGemma 2 is Google DeepMind's open, lightweight multimodal embedding model that maps text, code, images, video, and audio into a single shared embedding space.

Is EmbeddingGemma 2 open source?

The model weights are released under the Apache 2.0 license and are available on Hugging Face and Kaggle.

Can EmbeddingGemma 2 run on-device?

Yes. At 740M parameters it is designed for on-device use; the full multimodal model requires roughly 567MB of RAM on a Pixel 11 Pro, and text-only use about 191MB.

Google EmbeddingGemma 2

An open, lightweight multimodal embedding model

大语言模型

开发工具

访问

收录信息

HiCyou 收录记录

已收录

收录网站: blog.google
收录记录创建时间: 2026年10月13日
记录最后更新: 2026年10月13日

条目包含的信息

概述
4 项核心功能
3 个使用场景

此页面说明的是 HiCyou 的收录记录，不构成安全性、所有权或产品质量认证。

在提供敏感信息或购买前，请自行核查外部网站。

Google EmbeddingGemma 2 是什么

EmbeddingGemma 2 is Google DeepMind's open, lightweight multimodal embedding model. It maps text, code, images, video, and audio into a single shared embedding space, enabling cross-modal search from one query. At 740M parameters it is designed to run on-device, with weights released under the Apache 2.0 license on Hugging Face and Kaggle.

核心功能

Multimodal embeddings across text, code, images, video, and audio in one shared space

Lightweight 740M-parameter model that can run on-device

Open weights under the Apache 2.0 license, available on Hugging Face and Kaggle

Optimized for privacy-first use cases

使用场景

Cross-modal semantic search across text, images, video, and audio
On-device applications requiring local, privacy-preserving embeddings
Retrieval and search features built on open embedding models

创业公司为什么需要这个工具？

Startups can integrate an open, Apache 2.0-licensed multimodal embedding model without heavy infrastructure, enabling privacy-first, on-device search and retrieval features across multiple media types.