Hi there,
you already have documentation how to use it with datasette and with sqlite-utils - but where it would really shine, is using it with llm to really speed up embeddings search.
I'm pretty sure its possible, and probably easy once one knows how.
What do you think?
