I just tested the latest 2023 python version of USPEX and found that it does not work well for molecular crystals.
Sometimes, there exist (unphysical) structural overlap between neighboring molecules in the randomly generated structures. I am wondering whether this is a bug, or maybe it is just that the current version is not yet ready for molecular crystal work.
Thanks in advance for your inputs!
Best,
Peter