Google 发布 EmbeddingGemma 2 多模态嵌入模型
Google claims EmbeddingGemma 2 outperforms rival embedding models twice its size
Google 发布开源多模态嵌入模型 EmbeddingGemma 2，可将文本、图像、视频、音频和代码转为向量，参数量 740M，Google 称其在多模态嵌入基准上超过参数量两倍的竞品。该模型无需 API key 即可本地运行，浏览器中通过 WebGPU 单次查询约 20 至 70 毫秒，仅需约 191MB 内存，本地向量数据库存储最多减少六倍；纯文本任务可用 270M 参数版本。搭配 Gemma 4 等小型开源模型可离线运行 RAG 应用，权重已在 Hugging Face 和 Kaggle 发布，并附开发者指南与文档。
推荐理由
AI 评分 67
原文给出 740M 参数、191MB 内存和浏览器端 20-70ms 延迟等具体指标，可据此判断本地多模态嵌入的可用边界。
主题
正文
当前显示原文，中文全文译文尚未就绪。
Google released EmbeddingGemma 2, an open model that converts text, images, video, audio, and code into numerical vectors so similar content can be found and compared more easily. At 740 million parameters, Google says it's the most compact model of its kind and outperforms competing models up to twice its size on multimodal embedding benchmarks.
The model runs locally without an API key. Each query takes about 20 to 70 milliseconds via WebGPU in the browser. It needs only around 191 MB of RAM and cuts local vector database storage by up to six times. For text-only tasks, a 270-million-parameter version is enough.
Paired with small open models like Gemma 4 , EmbeddingGemma 2 can run offline RAG apps without sending data to external servers. The weights are available on Hugging Face and Kaggle , along with a developer guide and documentation .