Threadripper build for inference on dual 3090

[tensorworks]

Did a bit of testing and thought I'd post the result here.

Short version: vLLM behaved well for me, but only after I stopped ignoring token cost.

Nothing scientific here, just practical notes from a real setup.

I care more about stability than squeezing the last possible tokens/sec out of it.

[southpatch]

Thanks, that helps. That matches what I saw on 7900 XTX. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[turbosupply83]

That matches what I was seeing. That matches what I saw on GB200. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.