+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Hardware (https://www.datasculpting.ai/inference/forumdisplay.php?fid=4)
+--- Forum: GPUs (https://www.datasculpting.ai/inference/forumdisplay.php?fid=28)
+--- Thread: Used A6000 worth it now: anyone testing this? (/showthread.php?tid=340)
Used A6000 worth it now: anyone testing this? - syshouse - 09-18-2026
This keeps coming up in conversation so I figured I'd start a thread.
The more I work with local setups, the more it seems like tail latency decides everything downstream.
Curious whether that lines up with what other people are seeing.
Re: Used A6000 worth it now: anyone testing this? - vrambandit - 09-20-2026
I don't think that's the whole story. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.
Re: Used A6000 worth it now: anyone testing this? - sanevault - 09-20-2026
I hit almost the same issue last night.
Re: Used A6000 worth it now: anyone testing this? - paperrunner - 09-20-2026
that fixed it for me too.
Re: Used A6000 worth it now: anyone testing this? - rocmriley - 09-20-2026
that fixed it for me too.
Re: Used A6000 worth it now: anyone testing this? - cudapoint - 09-20-2026
Can you fill in a bit more context? Do you see the same thing with a smaller batch or shorter prompt?
Re: Used A6000 worth it now: anyone testing this? - paperrunner - 09-20-2026
Quote:Can you fill in a bit more context? Do you see the same thing with a smaller batch or shorter prompt?
A couple questions before I guess too hard. Are you testing this on 7900 XTX or something else?