+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Inference Engines (https://www.datasculpting.ai/inference/forumdisplay.php?fid=9)
+--- Forum: TensorRT-LLM (https://www.datasculpting.ai/inference/forumdisplay.php?fid=37)
+--- Thread: serving on Blackwell on dual 3090 (/showthread.php?tid=1023)
serving on Blackwell on dual 3090 - kernelgrid - 10-11-2026
Wanted to compare notes with people doing similar things.
I'm trying to decide whether to keep the current stack or change course. Main question is how people here are handling quant drift.
If you've tested something similar, what did you keep and what did you stop bothering with?
Re: serving on Blackwell on dual 3090 - northcraft90 - 10-12-2026
I'd start with the simpler stack.
Re: serving on Blackwell on dual 3090 - kernelmike - 10-12-2026
Thanks, that helps. I ended up in about the same place using vLLM. The biggest change for me was simplifying the stack instead of adding one more optimization layer.