+- 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 on mini PC (/showthread.php?tid=1038)
Pages:
1
2
4090 vs 5090 for 70B quants on mini PC - fusedhall - 10-12-2026
Posting this in case it saves someone else a little time.
If you were starting from scratch today for this use case, what would you choose first and what would you ignore?
I'm especially interested in mistakes worth avoiding.
I care more about stability than squeezing the last possible tokens/sec out of it.
Re: 4090 vs 5090 for 70B quants on mini PC - paperrunner - 10-12-2026
Not sure I agree there. 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.
Re: 4090 vs 5090 for 70B quants on mini PC - embersmith - 10-15-2026
Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher. The surprise was that memory pressure showed up before throughput collapsed.
Re: 4090 vs 5090 for 70B quants on mini PC - rocmriley - 10-15-2026
Appreciate the detail. That matches what I saw on 4090. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
Re: 4090 vs 5090 for 70B quants on mini PC - fusedhall - 10-15-2026
following this because I'm in the same boat.
Re: 4090 vs 5090 for 70B quants on mini PC - sanevault - 10-17-2026
I think the bottleneck is somewhere else. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Could be a workload difference, but I wouldn't generalize too fast from one setup.
Re: 4090 vs 5090 for 70B quants on mini PC - cudapoint - 10-18-2026
A couple questions before I guess too hard. Do you see the same thing with a smaller batch or shorter prompt?
Re: 4090 vs 5090 for 70B quants on mini PC - vrambandit - 10-18-2026
Ran a quick check on my box. My result was mixed: first run looked bad, second run looked normal. I expected the model choice to matter more than it did.
Re: 4090 vs 5090 for 70B quants on mini PC - cudapoint - 10-18-2026
Not sure I agree there. I've seen people blame batch size when the real issue was much more mundane. Worth testing the simpler explanation before adopting the more interesting one.
Re: 4090 vs 5090 for 70B quants on mini PC - sanevault - 10-18-2026
that fixed it for me too.