+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: LLM Models (https://www.datasculpting.ai/inference/forumdisplay.php?fid=12)
+--- Forum: Proprietary / Frontier (https://www.datasculpting.ai/inference/forumdisplay.php?fid=48)
+--- Thread: 8B vs 14B sweet spot: anyone testing this? (/showthread.php?tid=1456)
Re: 8B vs 14B sweet spot: anyone testing this? - cozywire - 10-18-2026
I'd approach it in a boring order. On 4090, the easiest win is often to back off the aggressive settings and see whether the problem survives.
8B vs 14B sweet spot: anyone testing this? - torchsmith25 - 10-18-2026
This keeps coming up in conversation so I figured I'd start a thread.
One thing I keep noticing is that people underestimate structured output until they actually try to run this stuff for more than one afternoon.
Wondering if this is just where the whole space is heading now.
Re: 8B vs 14B sweet spot: anyone testing this? - localpatch78 - 10-18-2026
A couple things I'd check first. If you're on PyTorch, 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.
Re: 8B vs 14B sweet spot: anyone testing this? - batchsize - 10-18-2026
Quote:I tested this tonight. On A6000 with TensorRT-LLM, I saw basically the same behavior. I expected the model choice to matter more than it did.
I think the bottleneck is somewhere else. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit. Could be a workload difference, but I wouldn't generalize too fast from one setup.
Re: 8B vs 14B sweet spot: anyone testing this? - quantnest - 10-18-2026
I tested this tonight. My result was mixed: first run looked bad, second run looked normal. The surprise was that memory pressure showed up before throughput collapsed.
Re: 8B vs 14B sweet spot: anyone testing this? - plainworks - 10-19-2026
Ran a quick check on my box. My result was mixed: first run looked bad, second run looked normal.
Re: 8B vs 14B sweet spot: anyone testing this? - southspark - 10-19-2026
Not sure I agree there. I get why that seems right, but on B200 the bottleneck felt elsewhere.
Re: 8B vs 14B sweet spot: anyone testing this? - fastscope - 10-19-2026
I tested this tonight. On A6000 with TensorRT-LLM, I saw basically the same behavior. I expected the model choice to matter more than it did.
Re: 8B vs 14B sweet spot: anyone testing this? - forgemixer - 10-19-2026
Tried to reproduce it just now. The issue went away once I changed one setting and restarted the stack cleanly. I expected the model choice to matter more than it did.