MinishLab/model2vec
A technique that compresses sentence transformer models by up to 50x to create lightweight static embedding models.
Not currently ranked — collecting fresh signals.
star history
Model2Vec turns any sentence transformer into a small, fast static embedding model. It produces distributional bag-of-words embeddings that maintain competitive performance while dramatically reducing model size. The technique enables efficient semantic search and retrieval without requiring full transformer inference.
Frequently asked
- What is MinishLab/model2vec?
- A technique that compresses sentence transformer models by up to 50x to create lightweight static embedding models.
- Is model2vec open source?
- Yes — MinishLab/model2vec is open source, released under the MIT license.
- What language is model2vec written in?
- MinishLab/model2vec is primarily written in Python.
- How popular is model2vec?
- MinishLab/model2vec has 2.2k stars on GitHub.
- Where can I find model2vec?
- MinishLab/model2vec is on GitHub at https://github.com/MinishLab/model2vec.