Stupid question from an outsider, but is it possible to grab a pretrained model (most likely one of the "camelids"), feed it your own data (I have about 1k docu

Stupid question from an outsider, but is it possible to grab a pretrained model (most likely one of the "camelids"), feed it your own data (I have about 1k documents that cannot leave my network) and use it for fast-but-sometimes-wrong information retrieval?