+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Optimization & Performance (https://www.datasculpting.ai/inference/forumdisplay.php?fid=10)
+--- Forum: Quantization (https://www.datasculpting.ai/inference/forumdisplay.php?fid=40)
+--- Thread: quant choices for 70B local (/showthread.php?tid=1462)
Pages:
1
2
Re: quant choices for 70B local - vrambandit - 10-19-2026
Good point. 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.
Re: quant choices for 70B local - vrambandit - 10-19-2026
Can you fill in a bit more context? Is this local-only or are you routing anything externally?
Re: quant choices for 70B local - torchshed - 10-19-2026
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.
Re: quant choices for 70B local - cudaworks - 10-19-2026
What exact setup are you running? What model, what runtime, and roughly what context size?
Re: quant choices for 70B local - vrambandit - 10-19-2026
What exact setup are you running? Do you see the same thing with a smaller batch or shorter prompt?