[arunsupe]

This implementation does not because the query has to be a word. One way to extend it to phrases is to average the vectors of each word in the phrase. Another way is to have a word2vec model that embeds phrases. (The large Google News model I think does have phrases, with spaces converted to underscore). But going from words to phrases opens up a whole new can of worms - how large a phrase to consider from the input stream etc. Plus, I don't think averaging words in the phrase is the same as learning the embedding for the phrase. Sentence embedding models are necessary for that, but they are far too slow for this use case as pointed out by others.

To summarize, this is a simple implementation that works for the simplest use case - semantic matching of words.

[vunderba]

Yep, it would almost be easier to:

- check to see if the search query contains more than one word

- spin up a more modestly sized LLM, such as mistral 7b

- ask the LLM to try to condense / find single word synonym for the user query

- send to sgrep