Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
SBC inference experiments for DeepSeek
#10
Quote:What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at prefix caching. Once

Good point. That matches what I saw on M3 Max. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
mostly local
Reply


Messages In This Thread
Re: SBC inference experiments for DeepSeek - by prompttinker44 - 09-15-2026, 10:41 PM

Forum Jump:


Users browsing this thread: 1 Guest(s)