Let us add this model: https://huggingface.co/sentence-transformers/paraphrase-MiniLM-L6-v2 to the list of supported Typesense models. As far as I understand: it is a small version ( 91 MB ) of this: https://huggingface.co/typesense/models/tree/main/paraphrase-multilingual-mpnet-base-v2 https://huggingface.co/sentence-transformers/paraphrase-multilingual-mpnet-base-v2 which is 1100 MB (and already supported by Typesense) The "paraphrase-MiniLM-L6-v2" trained on the same paraphrase datasets as the multilingual model, but only using the English portions. The difference in size is 10 times! P.S. I am user of such models, not a data scientist, so I can not do PR myself.. pls help!