Google claims EmbeddingGemma 2 outperforms rival embedding models twice its size
Google released EmbeddingGemma 2, an open model that converts text, images, video, audio, and code into numerical vectors so similar content can be found and compared, according to the company. At 740 million parameters, Google says it is the most compact model of its kind and outperforms competing models up to twice its size on multimodal embedding benchmarks.
On the Massive Text Embedding Benchmark (Code), EmbeddingGemma 2 scores 78.68, up nearly 10 points from its predecessor's 68.76, which Google says puts it on par with much larger models. The model runs locally without an API key, and each query takes about 20 to 70 milliseconds via WebGPU in the browser, according to the report.
It needs only around 191 MB of RAM and cuts local vector database storage by up to six times. For text-only tasks, a 270-million-parameter version is described as sufficient. Paired with small open models like Gemma 4, EmbeddingGemma 2 can run offline RAG apps without sending data to external servers.
The weights are available on Hugging Face and Kaggle, along with a developer guide and documentation. The performance and size claims are attributed to Google.
Based on reporting from the original publisher. Visit the source for full context and later updates.
Publisher excerpt
Google released EmbeddingGemma 2, an open model with 740 million parameters that converts text, images, video, audio, and code into vectors. It runs on-device, needs only about 191 MB of RAM, and outperforms some competing models twice its size, according to Google. Paired with a small open model like Gemma 4, it can run offline RAG apps without sending data to external servers. The article Google claims EmbeddingGemma 2 outperforms rival embedding models twice its size appeared first on The Decoder .