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 tokens/sec.

Happy to post more details if anyone wants them.

My use case is pretty ordinary: local chat, some document work, and occasional batch jobs. I keep promising myself this is the last rebuild.

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

One thing I keep noticing is that people underestimate consumer GPUs 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.

This is mostly for local work rather than public serving.

The part that surprised me was how much the boring details mattered more than the model choice.

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 TensorRT-LLM and A6000.

Simple and boring is a valid answer.

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

After a few runs on B200, I ended up with decent performance, but the setup only really made sense once I adjusted around tool calling.

Curious if anyone else landed in the same place.

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

I'm comparing two approaches here: keep it simple on A6000, 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.

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

After a few runs on A6000, I ended up with decent performance, but the setup only really made sense once I adjusted around KV cache pressure.

Happy to post more details if anyone wants them.

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.

This isn't revolutionary, but the combination of PyTorch and B200 felt surprisingly usable once I tuned around consumer GPUs.

Feel free to tear it apart if you see something dumb.

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, Threadripper box, with a lot of attention on prefix caching.

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

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

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

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

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

Welcome to the new Data Sculpting Inference Forum.

We're getting things organized and seeded so the board has a solid starting structure for local AI, inference, hardware, models, tools, and general discussion.

Please look around, post where it fits best, and let us know if there are sections you think should be added or adjusted.