A base-sized multilingual embedding model from the E5 family, designed for cross-lingual text retrieval and sentence similarity.