industry move that matters for local users

[deepseeked]

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

Just sharing a setup that ended up working reasonably well: Qwen, TensorRT-LLM, dual 3090, with a lot of attention on HBM constraints.

Maybe useful as a starting point for newer folks.

[routearch]

Ran a quick check on my box. My result was mixed: first run looked bad, second run looked normal. The surprise was that memory pressure showed up before throughput collapsed.