Text2vec Base Chinese Paraphrase is a sentence-transformers model hosted on Hugging Face for generating embeddings from Chinese text. It addresses the need to compute semantic similarities between Chinese sentences and phrases.
The model supports direct use through the sentence-transformers library by loading it with SentenceTransformer and encoding lists of sentences into embeddings. It can then calculate similarity matrices between those embeddings, as demonstrated in example code that processes four Chinese sentences and outputs a 4 by 4 similarity tensor. An alternative loading path uses the Transformers library with AutoTokenizer and AutoModel for direct access to the underlying architecture.
It belongs to the text2vec family and carries an apache-2.0 license. The model card identifies it under tasks including sentence similarity and feature extraction, with a base built on Chinese nli-zh-all and ernie components. Delivery occurs as a downloadable model repository on the Hugging Face platform, compatible with PyTorch and Safetensors formats.
No pricing information appears for this openly licensed model.
Text2vec Base Chinese Paraphrase is a Foundation models & chat product. It focuses on generating high-quality embeddings for Chinese sentences to compute semantic similarity. It is built as an open-source project for developers. Text2vec Base Chinese Paraphrase is open source under the Apache-2.0 license. The product ships for the web, the command line, and API.
shibing624 builds and maintains Text2vec Base Chinese Paraphrase, and the product first shipped in 2021. Development happens publicly on GitHub with 5k stars. Key capabilities include Sentence Embeddings, Semantic Similarity, and Chinese Language Support.
Latest indexed changes and source events
shibing624/text2vec-base-chinese-paraphrase verified by the PulseGate indexer
Other apps tracked under the same category.
