快讯
Google发布EmbeddingGemma 2：Apache 2.0轻量多模态嵌入模型，面向端侧检索
Google发布EmbeddingGemma 2，一款面向私有端侧搜索与检索的轻量多模态嵌入模型，采用Apache 2.0许可证。该740M模型将文本、代码、图像、视频和音频映射到同一共享空间，在MTEB Code上较EmbeddingGemma 1提升9.92分，量化后纯文本权重在Pixel 11 Pro上仅需约191MB活跃内存。
事件来源
查看原文
Google releases EmbeddingGemma 2, a lightweight multimodal embedding model built for private, on-device search and retrieval.📜 Apache 2.0. 🤖https://modelscope.ai/models/google/embeddinggemma-2 🏆 Delivers leading quality among sub-1B multimodal embedding models, including a 9.92-point gain over EmbeddingGemma 1 on MTEB Code. 🔎 Maps text, code, images, video, and audio into one shared space, enabling searches such as finding a video moment from a voice memo or searching audio with text. 📱 The 740M model uses as little as ~191MB active RAM for quantized text-only weights and ~567MB for the full multimodal model on a Pixel 11 Pro. 📦 Matryoshka embeddings shrink from 768d to 128d, reducing vector storage by up to 6×. 📚 Its 8K context handles up to 5.5 minutes of audio, 29 images, 58 video frames, or interleaved multimodal inputs.