+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Generative AI (https://www.datasculpting.ai/inference/forumdisplay.php?fid=17)
+--- Forum: Video (https://www.datasculpting.ai/inference/forumdisplay.php?fid=70)
+--- Thread: multimodal local stack that feels real with LM Studio (/showthread.php?tid=549)
Pages:
1
2
multimodal local stack that feels real with LM Studio - latentpeak22 - 09-29-2026
Maybe I'm missing something obvious.
My current setup is SGLang with Threadripper box, and I keep running into KV cache pressure.
Not looking for the perfect answer, mostly what worked in real use.
Re: multimodal local stack that feels real with LM Studio - phasespark - 10-01-2026
Quote:Maybe I'm missing something obvious.
A couple questions before I guess too hard. What model, what runtime, and roughly what context size?
Re: multimodal local stack that feels real with LM Studio - thetaember38 - 10-04-2026
That matches what I was seeing. I don't think your conclusion is far off.
Re: multimodal local stack that feels real with LM Studio - byteyard - 10-04-2026
Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher.
Re: multimodal local stack that feels real with LM Studio - latentpeak22 - 10-05-2026
I'd approach it in a boring order. On 7900 XTX, the easiest win is often to back off the aggressive settings and see whether the problem survives.
Re: multimodal local stack that feels real with LM Studio - turboarch95 - 10-06-2026
I tested this tonight. On dual 3090 with PyTorch, I saw basically the same behavior. The surprise was that memory pressure showed up before throughput collapsed.
Re: multimodal local stack that feels real with LM Studio - audiopath66 - 10-07-2026
Interesting, I hadn't tried that. I ended up in about the same place using LM Studio. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
Re: multimodal local stack that feels real with LM Studio - rasterhall - 10-07-2026
Quote:Interesting, I hadn't tried that. I ended up in about the same place using LM Studio. The biggest change for me was simplifying the stack instead of a
My first guess would be this: On B200, the easiest win is often to back off the aggressive settings and see whether the problem survives. Not glamorous advice, but repeatable baselines save a lot of time.
Re: multimodal local stack that feels real with LM Studio - cudapoint - 10-07-2026
A couple things I'd check first. If you're on Ollama, I'd confirm the version first and then reduce variables until the behavior is obvious. Once you know the plain path works, then add the faster path back in piece by piece.
Re: multimodal local stack that feels real with LM Studio - vectorshed43 - 10-07-2026
I don't think that's the whole story. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Worth testing the simpler explanation before adopting the more interesting one.