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 vLLM and GB200.

I'm especially interested in mistakes worth avoiding.

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: DeepSeek, llama.cpp, A6000, with a lot of attention on consumer GPUs.

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

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 vLLM and 4090.

Simple and boring is a valid answer.

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

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

It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with embeddings.

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.

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

Short version: MLX behaved well for me, but only after I stopped ignoring prefix caching.

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

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 NVFP4.

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

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

I've gone back and forth between Ollama and PyTorch for Kimi. The interesting difference for me has been MoE serving.

Interested in tradeoffs more than fan loyalty.

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

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

I'm comparing two approaches here: keep it simple on dual 3090, or squeeze more out of the stack and deal with the complexity.

Interested in tradeoffs more than fan loyalty.

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

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 long context.

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

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

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

Maybe useful as a starting point: PyTorch on dual 3090, keeping speculative decoding in mind from the start.

Maybe useful as a starting point for newer folks.