Inference › Optimization & Performance › Quantization v
quant choices for 70B local

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Linear Mode
quant choices for 70B local
torchshed Offline
Junior Member
**
Posts: 47
Threads: 6
Joined: Sep 2026
Reputation: 0
#13
10-19-2026, 03:53 AM
Did a small comparison this afternoon. On Threadripper box with TensorRT-LLM, I saw basically the same behavior. The main thing that moved was latency, not quality.
Find
Reply


Messages In This Thread
quant choices for 70B local - by fastbench - 10-18-2026, 10:43 PM
Re: quant choices for 70B local - by paperrunner - 10-18-2026, 08:44 PM
Re: quant choices for 70B local - by vrambandit - 10-18-2026, 09:08 PM
Re: quant choices for 70B local - by nanobrook - 10-18-2026, 09:39 PM
Re: quant choices for 70B local - by sparrowghost - 10-19-2026, 01:19 AM
Re: quant choices for 70B local - by paperrunner - 10-19-2026, 01:38 AM
Re: quant choices for 70B local - by fastbench - 10-19-2026, 01:46 AM
Re: quant choices for 70B local - by edgecoast - 10-19-2026, 01:52 AM
Re: quant choices for 70B local - by cacheworks - 10-19-2026, 01:24 AM
Re: quant choices for 70B local - by turboscope - 10-19-2026, 02:46 AM
Re: quant choices for 70B local - by vrambandit - 10-19-2026, 03:34 AM
Re: quant choices for 70B local - by vrambandit - 10-19-2026, 03:40 AM
Re: quant choices for 70B local - by torchshed - 10-19-2026, 03:53 AM
Re: quant choices for 70B local - by cudaworks - 10-19-2026, 04:24 AM
Re: quant choices for 70B local - by vrambandit - 10-19-2026, 05:16 AM

Forum Jump:


Users browsing this thread: 1 Guest(s)