Inference › Inference Engines › TensorRT-LLM v
quant support gaps on GB200

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Linear Mode
quant support gaps on GB200
kernelmike Offline
Member
***
Posts: 184
Threads: 23
Joined: Sep 2026
Reputation: 0
#13
08-30-2026, 10:07 PM
Quote:A couple questions before I guess too hard. What model, what runtime, and roughly what context size?

Good point. I don't think your conclusion is far off. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.
Mostly CUDA, sometimes wrong, usually measured
Find
Reply


Messages In This Thread
quant support gaps on GB200 - by routecrate - 08-26-2026, 08:31 PM
Re: quant support gaps on GB200 - by kernelmike - 08-26-2026, 09:59 PM
Re: quant support gaps on GB200 - by pixelcrate82 - 08-28-2026, 04:19 PM
Re: quant support gaps on GB200 - by vectorport - 08-27-2026, 12:20 AM
Re: quant support gaps on GB200 - by cozygrid12 - 08-27-2026, 08:41 AM
Re: quant support gaps on GB200 - by syshouse - 08-27-2026, 12:48 PM
Re: quant support gaps on GB200 - by kernelmike - 08-27-2026, 02:31 PM
Re: quant support gaps on GB200 - by turboyard94 - 08-27-2026, 03:30 PM
Re: quant support gaps on GB200 - by sysstation69 - 08-28-2026, 03:37 AM
Re: quant support gaps on GB200 - by kernelmike - 08-30-2026, 04:24 AM
Re: quant support gaps on GB200 - by northcraft90 - 08-30-2026, 08:32 AM
Re: quant support gaps on GB200 - by sysstation69 - 08-30-2026, 09:22 PM
Re: quant support gaps on GB200 - by kernelmike - 08-30-2026, 10:07 PM
Re: quant support gaps on GB200 - by labline - 08-30-2026, 10:50 PM
Re: quant support gaps on GB200 - by kernelpeak - 08-31-2026, 07:14 AM

Forum Jump:


Users browsing this thread: 1 Guest(s)