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

I'm comparing two approaches here: keep it simple on B200, or squeeze more out of the stack and deal with the complexity.

Trying to avoid rebuilding the whole thing twice.

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

Has anyone else run into this?

My current setup is TensorRT-LLM with A6000, and I keep running into prefix caching.

Not looking for the perfect answer, mostly what worked in real use. Turns out 'just one more test' was a lie.

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

I'm helping a small team set up a local workflow. We don't need a giant stack, but we do care about reranking.

Simple and boring is a valid answer.

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

Two paths both seem viable, but they break down differently once structured output becomes important.

Interested in tradeoffs more than fan loyalty.

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

I'm comparing two approaches here: keep it simple on M3 Max, or squeeze more out of the stack and deal with the complexity.

Trying to avoid rebuilding the whole thing twice. The logs were somehow calmer than I was.

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?

Simple and boring is a valid answer.

This is mostly for local work rather than public serving.

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

Short version: SGLang behaved well for me, but only after I stopped ignoring long context.

Curious if anyone else landed in the same place.

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

I'm helping a small team set up a local workflow. We don't need a giant stack, but we do care about quant drift.

Would rather learn from your scars than collect my own all at once.

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

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: Mixtral, PyTorch, 4090, with a lot of attention on long context.

Maybe useful as a starting point for newer folks.

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

I'm not trying to build a perfect architecture diagram. I just want a sane starting point around llama.cpp and dual 3090.

Would rather learn from your scars than collect my own all at once.