Pool spare GPU capacity to run LLMs at larger scale

 (github.com)

10 points | by i386 4 hours ago

2 comments

  • iwinux 3 hours ago
    You lost me on "spare GPU". I don't have any capable GPUs, let alone spare ones :)
  • vagrantJin 1 hour ago
    This is very promising, definitely looks more user friendly than exo. Can't wait to try it out.