+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Evaluation & Testing (https://www.datasculpting.ai/inference/forumdisplay.php?fid=16)
+--- Forum: Benchmarks (https://www.datasculpting.ai/inference/forumdisplay.php?fid=65)
+--- Thread: comparing models without fooling yourself: what are you seeing? (/showthread.php?tid=1427)
comparing models without fooling yourself: what are you seeing? - cudascope - 10-17-2026
Did a bit of testing and thought I'd post the result here.
I tried this with Phi using LM Studio on Threadripper box. 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.
Re: comparing models without fooling yourself: what are you seeing? - batchsize - 10-17-2026
Not sure I agree there. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Could be a workload difference, but I wouldn't generalize too fast from one setup.
There are a few different conversations hiding inside this thread, so being explicit about the goal helps.
Re: comparing models without fooling yourself: what are you seeing? - latentcraft - 10-18-2026
What worked for me was pretty simple. If you're on SGLang, 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.
Re: comparing models without fooling yourself: what are you seeing? - audionode8 - 10-18-2026
What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at workstation thermals.
Re: comparing models without fooling yourself: what are you seeing? - llm_wren - 10-19-2026
Interesting, I hadn't tried that. I don't think your conclusion is far off. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
Re: comparing models without fooling yourself: what are you seeing? - benchplease - 10-19-2026
Need a little more detail here. Are you testing this on 5090 or something else?
Re: comparing models without fooling yourself: what are you seeing? - thetapath - 10-19-2026
My first guess would be this: On 7900 XTX, the easiest win is often to back off the aggressive settings and see whether the problem survives.
Re: comparing models without fooling yourself: what are you seeing? - nanobrook - 10-19-2026
Good point. I don't think your conclusion is far off. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.
Re: comparing models without fooling yourself: what are you seeing? - cacheline - 10-19-2026
What exact setup are you running? Is this local-only or are you routing anything externally?