Inference › Inference Engines › Other Engines v
NVFP4 / INT4 tradeoffs

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Linear Mode
NVFP4 / INT4 tradeoffs
cachebench Offline
Member
***
Posts: 63
Threads: 13
Joined: Sep 2026
Reputation: 0
#4
10-02-2026, 04:35 PM
Maybe, but I had the opposite result. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Worth testing the simpler explanation before adopting the more interesting one.
Find
Reply


Messages In This Thread
NVFP4 / INT4 tradeoffs - by northhall88 - 09-28-2026, 03:08 AM
Re: NVFP4 / INT4 tradeoffs - by stackarch - 09-29-2026, 03:35 PM
Re: NVFP4 / INT4 tradeoffs - by visionforge - 09-29-2026, 04:02 PM
Re: NVFP4 / INT4 tradeoffs - by turboscope - 10-03-2026, 04:43 AM
Re: NVFP4 / INT4 tradeoffs - by cachebench - 10-02-2026, 04:35 PM
Re: NVFP4 / INT4 tradeoffs - by quietcraft - 10-06-2026, 05:26 AM
Re: NVFP4 / INT4 tradeoffs - by audiodesk - 10-09-2026, 05:53 AM
Re: NVFP4 / INT4 tradeoffs - by relaysupply90 - 10-10-2026, 06:39 AM
Re: NVFP4 / INT4 tradeoffs - by kernelhouse70 - 10-10-2026, 03:01 PM
Re: NVFP4 / INT4 tradeoffs - by rocmcrate26 - 10-10-2026, 03:15 PM
Re: NVFP4 / INT4 tradeoffs - by benchplease - 10-10-2026, 07:55 PM
Re: NVFP4 / INT4 tradeoffs - by visionforge - 10-10-2026, 08:35 PM
Re: NVFP4 / INT4 tradeoffs - by marbleyard32 - 10-11-2026, 09:07 PM

Forum Jump:


Users browsing this thread: 2 Guest(s)