As I pointed in other comments, the current implementation does not focus on the scale problem. However, using the 'scores' attribute of the 'find_duplicates' function, one could obtain the hamming distance/cosine similarity and then use that to sort. For more, please refer the docs.

The API does not seem to support this, but it should be easy to hack this (return not only a list of duplicates but the actual distance to the target image along with it, then sort results by distance).

The api already supports returning the hamming distances/cosine similarities along with the duplicate file list which can be used to sort the files. Please refer the docs for 'find_duplicates' function for more.