Has anyone else run into this?

I'm trying to decide whether to keep the current stack or change course. Main question is how people here are handling small local stacks.

If you've tested something similar, what did you keep and what did you stop bothering with?

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

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 embeddings.

I'm especially interested in mistakes worth avoiding.

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

After a few runs on 5090, I ended up with decent performance, but the setup only really made sense once I adjusted around structured output.

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

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

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

Trying to avoid rebuilding the whole thing twice.

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 consumer GPUs.

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

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

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

The more I work with local setups, the more it seems like KV cache pressure decides everything downstream.

Curious whether that lines up with what other people are seeing.

My use case is pretty ordinary: local chat, some document work, and occasional batch jobs.

Has anyone else run into this?

My current setup is vLLM with 7900 XTX, and I keep running into token cost.

If you've tested something similar, what did you keep and what did you stop bothering with?

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

I've gone back and forth between llama.cpp and MLX for FLUX. The interesting difference for me has been INT4.

Trying to avoid rebuilding the whole thing twice.

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

After a few runs on GB200, I ended up with decent performance, but the setup only really made sense once I adjusted around tool calling.

Happy to post more details if anyone wants them.

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: LM Studio behaved well for me, but only after I stopped ignoring speculative decoding.

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