you are viewing a single comment's thread
view the rest of the comments
[–] 5 points 4 days ago

Local models are astoundingly slow compared to online ones and you can't really run anything else while they are doing inference.

  • source
  • parent