Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
throughput per watt reality check with llama.cpp
#5
I think the bottleneck is somewhere else. I get why that seems right, but on GB200 the bottleneck felt elsewhere. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.
rack in the garage • local models everywhere
Reply


Messages In This Thread
Re: throughput per watt reality check with llama.cpp - by homelabhenry - 09-09-2026, 04:07 PM

Forum Jump:


Users browsing this thread: 1 Guest(s)