The community member is looking for a free and open-source software (FOSS) vectorizer to store things in vector databases, without using the OpenAI embeddings API. Another community member suggests using any embedding model from Huggingface, which can run locally. The community members discuss that the embedding models are typically not very large in size, and that the solution will run completely on-premises, except for downloading the model weights.
Is there a FOSS vectorizer to store things in the vectorDBs? I don't want to use the OpenAI embeddings API, looking for something that can be run entirely local.