Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Instinct pricing thoughts
#11
Quote:Appreciate the detail. The pattern around MoE serving has been pretty similar for me. Worth rechecking your defaults though, because one bad assumptio

I don't think that's the whole story. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model.
Reply
#12
worth checking the driver first.
Reply
#13
That matches what I was seeing. I ended up in about the same place using vLLM. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
7900 XTX / ROCm / somehow still optimistic
Reply
#14
Appreciate the detail. The pattern around HBM constraints has been pretty similar for me. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
team red, team patient
Reply
#15
Quote:Appreciate the detail. The pattern around HBM constraints has been pretty similar for me. The biggest change for me was simplifying the stack instead

Thanks, that helps. I don't think your conclusion is far off. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

There are a few different conversations hiding inside this thread, so being explicit about the goal helps.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)