+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Hardware (https://www.datasculpting.ai/inference/forumdisplay.php?fid=4)
+--- Forum: GPUs (https://www.datasculpting.ai/inference/forumdisplay.php?fid=28)
+--- Thread: 4090 vs 5090 for 70B quants: anyone testing this? (/showthread.php?tid=1161)
4090 vs 5090 for 70B quants: anyone testing this? - nanobrook - 10-13-2026
Posting this in case it saves someone else a little time.
I'm not trying to build a perfect architecture diagram. I just want a sane starting point around Ollama and M3 Max.
Simple and boring is a valid answer.
My use case is pretty ordinary: local chat, some document work, and occasional batch jobs. The logs were somehow calmer than I was.
Re: 4090 vs 5090 for 70B quants: anyone testing this? - thetaember38 - 10-13-2026
What worked for me was pretty simple. Before rebuilding anything, I'd test one clean baseline with Llama and compare from there.
Re: 4090 vs 5090 for 70B quants: anyone testing this? - vrambandit - 10-14-2026
Did a small comparison this afternoon. My result was mixed: first run looked bad, second run looked normal. The main thing that moved was latency, not quality.