Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Tensor cores actually helping here
#11
I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at workstation thermals. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
still benchmarking everything
Reply
#12
What exact setup are you running? What model, what runtime, and roughly what context size?
Reply
#13
worth checking the driver first.
Mostly CUDA, sometimes wrong, usually measured
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)