MODELS

sentence-transformers/paraphrase-multilingual-mpnet-base-v2

by sentence-transformers

modellead-sourcehugging-face-popular-modelssource:github.comsentence-transformershugging-facesentence-similaritymultilingualembeddings

Visit Website Documentation GitHub

Overview

Multilingual Sentence Transformers model for semantic sentence similarity.

Details

sentence-transformers/paraphrase-multilingual-mpnet-base-v2 is listed on Hugging Face as a sentence-similarity model. The official Sentence Transformers documentation lists it as a multilingual semantic-similarity model trained on parallel data for 50+ languages. The related Sentence Transformers GitHub repository describes the framework as a way to compute embeddings with pretrained Sentence Transformer models and links to the Hugging Face model collection and SBERT documentation.

When to Use

Use when you need sentence-level semantic similarity with multilingual coverage. Use with the Sentence Transformers framework when you want embeddings from a pretrained Sentence Transformer model. Use when comparing or clustering paraphrases across supported languages is part of the workflow.

Getting Started

Open the model page on Hugging Face: https://huggingface.co/sentence-transformers/paraphrase-multilingual-mpnet-base-v2
Review the official Sentence Transformers pretrained-model documentation for this model’s multilingual semantic-similarity context.
Use the Sentence Transformers GitHub repository and documentation to install and run the framework with pretrained models.
Run a small sentence-similarity test on representative multilingual examples before production use.

Key Features

•Listed by Hugging Face under sentence-similarity.
•Documented by Sentence Transformers as a multilingual semantic-similarity model.
•Trained on parallel data for 50+ languages according to the official Sentence Transformers documentation.
•Designed for use with the Sentence Transformers embedding framework.

Capabilities

•sentence similarity
•multilingual semantic similarity
•sentence embeddings
•pretrained embedding model

Last updated Jun 5, 2026