>>513877731
>Any ideas about how to give them proper filenames?
I used metadata. Artist name, date, description, type of mounting/size depending on filename constraints (256 chars).
>As soon as we figure out how to get an LLM to do that for us we can just concentrate on filtering and archiving the best stuff.
GREAT idea.

Here's the thing tho... I managed to extract cached image from Amazon web index. Problem is that many are 400px wide. Dorotheum has a nasty habit of removing high resolution images fter a month so these high-res images are forever lost unfortunately. I did retrieve all I could.
Smaller 400px images are there since that might be all we have.
Cached directory has only filename names because Dorotheum removed them completely off their site and Amazon's web index (which goes back a decade) still has them.