Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
framework overhead in real deployments
#11
My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at INT4. Once you know the plain path works, then add the faster path back in piece by piece.
Mostly CUDA, sometimes wrong, usually measured
Reply


Messages In This Thread
Re: framework overhead in real deployments - by kernelmike - 09-23-2026, 06:58 PM

Forum Jump:


Users browsing this thread: 1 Guest(s)