Google releases EmbeddingGemma 2, a 744M multimodal embedding model
Modular encoders let developers deploy only the weights a workload needs, cutting memory cost without leaving the shared vector space. That makes it a practical on-device choice for agent memory and retrieval over mixed media, where running a larger multimodal embedder locally is usually infeasible. Teams comparing against larger closed embedding APIs can benchmark the int4 build first.
Google has released EmbeddingGemma 2, an open 744M-parameter multimodal embedding model that handles text, code, images, video, documents and audio in a single vector space. Each modality has its own encoder feeding into a shared backbone, allowing selective loading: text-only runs at 271M parameters, vision-language at 439M, and audio-language at 577M. The model scores 61.36 on MTEB Multilingual v2 and 68.46 on MTEB English v2, beating specialist and larger models across seven benchmark suites. The int4 quantised version loses about 0.4 points on MTEB Multilingual, and both the vision token budget and embedding dimensions can be reduced further.
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
