Google 开源 744M 多模态嵌入模型 EmbeddingGemma 2
Google 开源了个 744M 的多模态嵌入模型，文本图片音频视频全丢一个向量空间，参数还能按模态裁剪，做本地 RAG 或智能体记忆可以直接试试。
Google 发布开源嵌入模型 EmbeddingGemma 2，参数量 744M，支持文本、图像、视频、文档和音频。该模型在覆盖六类模态的七套基准上超过体积大 2 到 12 倍的专门模型。它在 MTEB Multilingual v2 得分 61.36，MTEB English v2 得分 68.46，int4 量化版本在 MTEB Multilingual 上仅损失约 0.4 分。各模态编码器共享一个主干，可按需加载，纯文本 271M、视觉语言 439M、音频语言 577M。还支持降低视觉 token 预算和截断嵌入维度。
Banger report from Google.
They just released EmbeddingGemma 2, an open 744M multimodal embedding model.
It beats specialist models and models 2 to 12 times larger across seven benchmark suites covering text, code, images, video, documents, and audio.
Text, images, audio, video, and mixed inputs all go into one vector space. Each modality has its own encoder feeding into a shared backbone, so you can load only what you need. Text-only runs at 271M parameters, vision-language at 439M, and audio-language at 577M.
It scores 61.36 on MTEB Multilingual v2 and 68.46 on MTEB English v2. The int4 version loses about 0.4 points on MTEB Multilingual.
You can also lower the vision token budget and truncate the embedding dimensions.
For agent memory and retrieval over mixed media, this small open model is a great on-device option.
https://t.co/tVBZFPjwaV
Chat with Paper: https://t.co/pdaNTOY37g
