Inference › Drivers / Frameworks / Toolkits › PyTorch / JAX / Other Frameworks v
framework overhead in real deployments

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Linear Mode
framework overhead in real deployments
cudaworks Offline
Junior Member
**
Posts: 23
Threads: 3
Joined: Sep 2026
Reputation: 0
#9
09-27-2026, 03:43 PM
Quote:Thanks, that helps. That matches what I saw on M3 Max.

A couple questions before I guess too hard. Do you see the same thing with a smaller batch or shorter prompt?
Find
Reply


Messages In This Thread
Re: framework overhead in real deployments - by cudaworks - 09-27-2026, 03:43 PM

Forum Jump:


Users browsing this thread: 1 Guest(s)