Inference Arena – new benchmark of local inference and training

2 comments

• $Alexzoofficial 3 days ago

  [flagged]

  [-]
  • $kvark 3 days ago

    Thank you for support!

    Part of the problem with timing variety is frameworks not always picking the right gpu/backend.

    If you want to inspect or tweak the setup, be my guest at https://github.com/kvark/inferena