+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Training & Customization (https://www.datasculpting.ai/inference/forumdisplay.php?fid=13)
+--- Forum: Distillation (https://www.datasculpting.ai/inference/forumdisplay.php?fid=55)
+--- Thread: when RAG was overkill (/showthread.php?tid=150)
Pages:
1
2
Re: when RAG was overkill - quantnest - 09-13-2026
A couple things I'd check first. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at HBM constraints. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
Re: when RAG was overkill - vramforge - 09-13-2026
My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Phi and compare from there. Once you know the plain path works, then add the faster path back in piece by piece.
I'm happy to test one more variant if you want a second data point.
Re: when RAG was overkill - kernelmike - 09-14-2026
Quote:My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Phi and compare from there. Once you know the plain path wo
What exact setup are you running? Do you see the same thing with a smaller batch or shorter prompt?
Re: when RAG was overkill - macropoint - 09-15-2026
I'd approach it in a boring order. On M3 Max, the easiest win is often to back off the aggressive settings and see whether the problem survives.
Re: when RAG was overkill - oxideworks59 - 09-16-2026
Appreciate the detail. 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.