Spent some time on this today and wanted to compare notes.

I'm comparing two approaches here: keep it simple on Threadripper box, or squeeze more out of the stack and deal with the complexity.

Interested in tradeoffs more than fan loyalty.

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: Phi, SGLang, A6000, with a lot of attention on token cost.

If someone wants the exact settings I can clean them up and post them.

This is mostly for local work rather than public serving.

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

After a few runs on Threadripper box, I ended up with decent performance, but the setup only really made sense once I adjusted around prefix caching.

Curious if anyone else landed in the same place.

Posting this in case it saves someone else a little time.

If you were starting from scratch today for this use case, what would you choose first and what would you ignore?

I'm especially interested in mistakes worth avoiding.

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

I tried this with Qwen using PyTorch on mini PC. The headline result was better than I expected, but quant drift still ended up being the limiting factor.

Happy to post more details if anyone wants them.

This keeps coming up in conversation so I figured I'd start a thread.

One thing I keep noticing is that people underestimate tool calling until they actually try to run this stuff for more than one afternoon.

Maybe obvious to some of you, but it clicked for me this week. At this point the real bottleneck is my patience.

Wanted to compare notes with people doing similar things.

I'm mostly working with Phi in Ollama on M3 Max. The part I'm trying to understand is prefill/decode split.

Not looking for the perfect answer, mostly what worked in real use.

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

After a few runs on dual 3090, I ended up with decent performance, but the setup only really made sense once I adjusted around router policy.

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

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: FLUX, TensorRT-LLM, mini PC, with a lot of attention on tail latency.

Maybe useful as a starting point for newer folks.

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

After a few runs on Threadripper box, I ended up with decent performance, but the setup only really made sense once I adjusted around tail latency.

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

This is mostly for local work rather than public serving.

If I had to summarize the lesson so far, it would be that clean setup and observability beat clever guesses.