The model I'm using for QA info extraction is too heavy


I’m trying to use this BERT-large model for some information extraction through question answering.
I’m just using it, out of the box, without any fitting.
However, it seems that this model is too big for my resources.

I’m looking for:

  1. alternative smaller models for the same task in the transformers package.
  2. techniques to reduce the weight of this model on my resources.