Google's small open embedding model that maps text, code, images, video and audio into a shared embedding space.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
EmbeddingGemma 2 is an embedding model in Google's Gemma family. It extends the original text-only EmbeddingGemma beyond text, mapping code, images, video and audio into a single shared embedding space, so content of different types can be compared and retrieved together.
According to a creator summary, the model is under a billion parameters and is small enough to run on phones, which suits on-device retrieval and semantic search. It is released as open weights under the Apache 2.0 license and is reported to be available on Hugging Face. Official details such as benchmarks, embedding dimensions and context lengths were not in the provided material.
Developers can use it through the Gemma documentation and ecosystem integrations, or download the weights and run them locally. It outputs vectors rather than generated content, so it is typically paired with a vector database or a generative model in RAG and search pipelines.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Small open multimodal embedding model for text, image, audio and video, under a billion parameters, running on phones. Apache 2 license and available on Hugging Face.
Closest neighbours by meaning, not just by category.
Model · Google
Google's 2B-parameter Gemma 2 instruction-tuned model, repackaged as a text encoder for PixelDiT and PiD workflows in ComfyUI.
Model · Google
Google's open-weights 12B encoder-free multimodal model in the Gemma 4 family, taking text, image and audio input.
Model · Google
Google's open-weights 12B multimodal (text and image input) instruction-tuned model from the Gemma 3 family.
Model · Google
Google DeepMind's family of open-weights models built from Gemini technology, from phone-sized to laptop-class variants.
Model · NVIDIA
NVIDIA family of multilingual text embedding models (8B and 1B) for retrieval, RAG and coding agents.
Model · AIDC-AI
Open multimodal embedding models from AIDC-AI that map text, images, video and documents into a shared vector space.
