This is a Transformers.js-compatible version of the paraphrase-multilingual-MiniLM-L12-v2 model hosted on Hugging Face. It generates high-quality sentence embeddings for over 50 languages, enabling semantic similarity, paraphrase detection, and retrieval-augmented generation applications. Developers can run it locally in browsers or Node.js without server-side dependencies.
In the Embeddings & retrieval space, Paraphrase Multilingual MiniLM L12 takes a focused approach. It focuses on converting multilingual text into dense vector embeddings for semantic search and similarity tasks. Paraphrase Multilingual MiniLM L12 is an open-source project aimed at developers. The project is open source (Open Source). It ships for the web, the command line, and API.
Xenova builds and maintains Paraphrase Multilingual MiniLM L12, and it first shipped in 2023. Among its 3 catalogued features are Feature Extraction, Sentence Embeddings, and Multilingual Support.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do