what are you all running this week

[marbletrack]

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

I tried this with Qwen using TensorRT-LLM on A6000. The headline result was better than I expected, but consumer GPUs still ended up being the limiting factor.

Happy to post more details if anyone wants them.

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

[vectorshed43]

Appreciate the detail. The pattern around tail latency has been pretty similar for me. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[thetagrid15]

Can you fill in a bit more context? Is this local-only or are you routing anything externally?

[marbletrack]

A couple things I'd check first. If you're on LM Studio, I'd confirm the version first and then reduce variables until the behavior is obvious. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

If you post your exact settings, people here can probably narrow it down pretty fast.

[thetagrid15]

My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Qwen and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[ggufwire22]

Tried to reproduce it just now. My result was mixed: first run looked bad, second run looked normal. The surprise was that memory pressure showed up before throughput collapsed.

[tensorkite]

Ran a quick check on my box. My result was mixed: first run looked bad, second run looked normal.

[oxidemixer]

Ran a quick check on my box. My result was mixed: first run looked bad, second run looked normal.

following this because I'm in the same boat.

[tensorkite]

Good point. I ended up in about the same place using MLX.

[quiettoken]

Interesting, I hadn't tried that. I ended up in about the same place using Ollama. The biggest change for me was simplifying the stack instead of adding one more optimization layer.