+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Hardware (https://www.datasculpting.ai/inference/forumdisplay.php?fid=4)
+--- Forum: Systems & Builds (https://www.datasculpting.ai/inference/forumdisplay.php?fid=29)
+--- Thread: rackmount vs tower for homelab inference: worth it? (/showthread.php?tid=889)
Pages:
1
2
Re: rackmount vs tower for homelab inference: worth it? - homelabhenry - 10-10-2026
I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at long context. Once you know the plain path works, then add the faster path back in piece by piece.
Re: rackmount vs tower for homelab inference: worth it? - tensorworks - 10-10-2026
A couple things I'd check first. If you're on PyTorch, I'd confirm the version first and then reduce variables until the behavior is obvious. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
Re: rackmount vs tower for homelab inference: worth it? - batchwire - 10-11-2026
I tested this tonight. My result was mixed: first run looked bad, second run looked normal.
Re: rackmount vs tower for homelab inference: worth it? - macroforge - 10-12-2026
I'd start with the simpler stack.
Re: rackmount vs tower for homelab inference: worth it? - kernelhouse70 - 10-14-2026
Good point. I ended up in about the same place using llama.cpp. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.
Re: rackmount vs tower for homelab inference: worth it? - homelabhenry - 10-15-2026
Thanks, that helps. 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.