Selecting the Embedding Model

We continue to utilize this embedding model. It’s important to note that the dense vector space of this model has a dimensionality of 384, which must be specified when setting up the vector database with Supabase.

from langchain_huggingface.embeddings import HuggingFaceEmbeddings
embedding = HuggingFaceEmbeddings(
    model_name="sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2"
)

You can find the details of “dimensional dense vector space” from Hugging Face at https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2