All Tools
M
RAGFreeOpen Source
MODEL2VEC
Distill sentence transformers into static embeddings
MIT
ABOUT
Sentence transformers are accurate but heavy to serve for every query. Model2Vec distills them into static token embeddings that average into a sentence vector, cutting latency and GPU cost while keeping enough quality for search, clustering, and classification.
INSTALL
pip install model2vecINTEGRATION GUIDE
1. Serve fast static embeddings for RAG retrieval without a GPU
2. Distill a domain sentence transformer into a tiny production model
3. Cluster or classify large document sets cheaper than full transformers
4. Swap a slow embedding endpoint for Model2Vec in a latency-sensitive search path
TAGS
pythonembeddingsragdistillationsentence-transformersopen-source