quiet dual-GPU build advice

[latentpeak22]

Thought I'd post a small recipe because a lot of threads ask similar things.

Maybe useful as a starting point: TensorRT-LLM on mini PC, keeping long context in mind from the start.

Maybe useful as a starting point for newer folks.

[tensorworks]

Did a small comparison this afternoon. The issue went away once I changed one setting and restarted the stack cleanly. I expected the model choice to matter more than it did.

[bytegarden73]

Did a small comparison this afternoon. The issue went away once I changed one setting and restarted the stack cleanly. I expected the model choice to

I hit almost the same issue last night.

[plainscope50]

Good point. The pattern around prefix caching has been pretty similar for me. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[labnest]

Maybe, but I had the opposite result. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit. Worth testing the simpler explanation before adopting the more interesting one.

[mellowstream]

A couple questions before I guess too hard. Do you see the same thing with a smaller batch or shorter prompt?

[macrotinker]

I'd approach it in a boring order. On B200, the easiest win is often to back off the aggressive settings and see whether the problem survives. Once you know the plain path works, then add the faster path back in piece by piece.

If you post your exact settings, people here can probably narrow it down pretty fast. The logs were somehow calmer than I was.

[nanobrook]

My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Mixtral and compare from there. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[bytegarden73]

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Qwen and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[kernelhouse70]

subscribed, curious where this lands.