Why download large model each time as local machine only need once?

#1
by Awiny - opened

In this code, I need to load some big models.

However, in local machine, with model initialize, the code only download big models once.

However, in this space, when I inference, it download big models each time, make the inference very very slow.

If the memory is full?

solved

Awiny changed discussion status to closed

Sign up or log in to comment