Sentence Transformers · Embedding
Paraphrase Multilingual MPNet v2
Sentence TransformersEmbeddingParaphrase familyOpen weights. The weights are publicly available in a Hugging Face repo.
Paraphrase Multilingual MPNet v2 is a sentence-transformer embedding model producing 768-dimensional vectors for multilingual semantic search and similarity.
Evidence and sources
🤗 sentence-transformers/paraphrase-multilingual-mpnet-base-v2llminfo IQ 84.3, ranked among embedding models. From 137 benchmarks. Plenty of data behind it.How llminfo IQ is computedReleased 27 Aug 2019
Sources: EURouter data, llminfo researchPrimary data from EURouter
from llminfo research: release date
Tags from its sources:embeddingmultilingual
Advantages and trade-offs
Compared with 10 embedding models, picked mainly by popularity and IQ
Trade-offs
Fewer provider choices
Among the weaker models in its class
Input & output
as stated by the model's providersText
Embeddings
Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works