image generation model locals actually use

[routecraft]

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

Short version: llama.cpp behaved well for me, but only after I stopped ignoring quant drift.

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

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

[cobalttinker]

My first guess would be this: If you're on TensorRT-LLM, I'd confirm the version first and then reduce variables until the behavior is obvious. Not glamorous advice, but repeatable baselines save a lot of time.

[amberpatch25]

Thanks, that helps. That matches what I saw on GB200. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[oxideworks59]

I'd start with the simpler stack.

[fastlab]

Interesting, I hadn't tried that. I ended up in about the same place using LM Studio. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[vramghost]

Interesting, I hadn't tried that. I ended up in about the same place using LM Studio. Worth rechecking your defaults though, because one bad assumptio

What worked for me was pretty simple. On M3 Max, the easiest win is often to back off the aggressive settings and see whether the problem survives. Once you know the plain path works, then add the faster path back in piece by piece.

[batchyard]

What worked for me was pretty simple. On M3 Max, the easiest win is often to back off the aggressive settings and see whether the problem survives. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.