Welcome EmbeddingGemma, Google's new efficient embedding model

Today, Google releases EmbeddingGemma, a state-of-the-art multilingual embedding model perfect for on-device use cases. Designed for speed and efficiency, the model features a compact size of 308M parameters and a 2K context window, unlocking new possibilities for mobile RAG pipelines, agents, and more. EmbeddingGemma is trained to support over 100 languages and is the highest-ranking text-only multilingual embedding model under 500M on the Massive Text Embedding Benchmark (MTEB) at the time of writing.

Text embeddings have become the backbone of modern natural‑language applications, turning words, sentences, and documents into dense vectors that capture meaning, sentiment, and intent. These vectors enable fast similarity search, clustering, classification, and retrieval across massive corpora, powering everything from recommendation engines and semantic search to retrieval-augmented generation and code‑search tools. Embedding models that calculate these embeddings are widely used, with well over 200 million monthly downloads on Hugging Face.

Building on this foundation, Google DeepMind’s EmbeddingGemma arrives as the newest, most capable small multilingual embedding model yet. With just 308M parameters, a 2k‑token context window, and support for over 100 languages, EmbeddingGemma delivers state‑of‑the‑art performance on the Massive Multilingual Text Embedding Benchmark (MMTEB) while staying under 200 MB of RAM when quantized.

The various design choices result in a very practical, open-source tool for computing high-quality multilingual embeddings on everyday devices.

EmbeddingGemma builds on the Gemma3 transformers backbone, but modified to use bi-directional attention instead of causal (one-way) attention. This means that earlier tokens in the sequence can attend to later tokens, effectively turning the architecture from a decoder into an encoder. Encoder models can outperform LLMs, which are decoders, on embedding tasks like retrieval. With this backbone, the model can process a sizable 2048 tokens at once, sufficient for typical retrieval inputs.

Beyond the new Gemma3-based encoder backbone, which produces token embeddings, a mean pooling layer converts these token embeddings into text embeddings. Lastly, two dense layers transform the text embeddings into their final form, a 768-dimensional vector. The model has been trained with Matryoshka Representation Learning (MRL), allowing you to truncate the 768‑dimensional output to 512, 256, or 128 dimensions on demand.

The model has been trained using a carefully curated, multilingual corpus totalling approximately 320 billion tokens. EmbeddingGemma was benchmarked on the MMTEB (Multilingual, v2) and MTEB (English, v2) suites, where it consistently beats comparable baselines while keeping a very small memory footprint.

EmbeddingGemma is integrated with many popular tools, including Sentence Transformers, LangChain, LlamaIndex, Haystack, and txtai. For production deployments, you can use Text Embeddings Inference (TEI) or Transformers.js for web applications.