Google releases EmbeddingGemma 2, an open on-device model for multimodal search
Google DeepMind released EmbeddingGemma 2, a 740-million-parameter open model that maps text, code, images, audio and video into one shared embedding space. It is released under the Apache 2.0 license.
Google announced EmbeddingGemma 2 on October 6. The model extends its predecessor beyond text, placing code, images, video and audio in a single embedding space so that, for example, a text query can search audio recordings.
Google says the model is built on the Gemma 4 architecture, has 740 million parameters to suit on-device use, and is released under the commercially permissive Apache 2.0 license. The company says the first EmbeddingGemma passed 20 million downloads.