rackmount vs tower for homelab inference: worth it?

[plainscope50]

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 speculative decoding decides everything downstream.

Wondering if this is just where the whole space is heading now.

For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.

[emberpeak]

subscribed, curious where this lands.

[rocmplane]

Tried to reproduce it just now. The issue went away once I changed one setting and restarted the stack cleanly. The surprise was that memory pressure showed up before throughput collapsed.

[cozygrid12]

I think the bottleneck is somewhere else. I've seen people blame quant drift when the real issue was much more mundane. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.

[tensorport]

Thanks, that helps. I don't think your conclusion is far off.

[cozygrid12]

Thanks, that helps. I don't think your conclusion is far off.

subscribed, curious where this lands.

[labnest]

Thanks, that helps. I don't think your conclusion is far off.

Tried to reproduce it just now. My result was mixed: first run looked bad, second run looked normal. I expected the model choice to matter more than it did.

[batchwire]

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Llama and compare from there.

[homelabhenry]

A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. Once you know the plain path works, then add the faster path back in piece by piece.

[latentpeak22]

What exact setup are you running? Are you testing this on GB200 or something else?