Inference › Optimization & Performance › Benchmarks v
QLoRA resource reality check

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Linear Mode
QLoRA resource reality check
cudascope Offline
Member
***
Posts: 50
Threads: 2
Joined: Sep 2026
Reputation: 0
#10
10-06-2026, 08:25 PM
Maybe, but I had the opposite result. I get why that seems right, but on GB200 the bottleneck felt elsewhere. Could be a workload difference, but I wouldn't generalize too fast from one setup.
learning in public
Find
Reply


Messages In This Thread
QLoRA resource reality check - by cacheline - 10-03-2026, 04:04 PM
Re: QLoRA resource reality check - by batchsize - 10-03-2026, 05:11 PM
Re: QLoRA resource reality check - by quiethouse - 10-05-2026, 06:54 PM
Re: QLoRA resource reality check - by benchplease - 10-08-2026, 01:15 PM
Re: QLoRA resource reality check - by batchsize - 10-04-2026, 02:21 AM
Re: QLoRA resource reality check - by nanobrook - 10-05-2026, 07:14 PM
Re: QLoRA resource reality check - by vectorpoint - 10-06-2026, 04:12 PM
Re: QLoRA resource reality check - by cudascope - 10-06-2026, 08:25 PM
Re: QLoRA resource reality check - by cacheline - 10-07-2026, 08:43 PM
Re: QLoRA resource reality check - by localtinker - 10-08-2026, 12:56 AM
Re: QLoRA resource reality check - by cacheline - 10-08-2026, 09:57 PM
Re: QLoRA resource reality check - by routearch - 10-09-2026, 10:28 PM

Forum Jump:


Users browsing this thread: 1 Guest(s)