10-10-2026, 06:29 AM
I tested this tonight. On 4090 with LM Studio, I saw basically the same behavior. I expected the model choice to matter more than it did.
Mostly CUDA, sometimes wrong, usually measured
Inference
›
Optimization & Performance
›
Performance Tuning
consumer GPU serving economics for Gemma
consumer GPU serving economics for Gemma
|
consumer GPU serving economics for Gemma
|
|
« Next Oldest | Next Newest »
|
| Messages In This Thread |
|
consumer GPU serving economics for Gemma - by papermixer4 - 10-08-2026, 04:04 AM
Re: consumer GPU serving economics for Gemma - by sparrowshed - 10-08-2026, 04:29 AM
Re: consumer GPU serving economics for Gemma - by rocmpath - 10-08-2026, 04:39 PM
Re: consumer GPU serving economics for Gemma - by signalcraft14 - 10-10-2026, 04:57 AM
Re: consumer GPU serving economics for Gemma - by kernelmike - 10-10-2026, 06:29 AM
Re: consumer GPU serving economics for Gemma - by fastscope - 10-10-2026, 06:42 PM
Re: consumer GPU serving economics for Gemma - by edgeriver - 10-12-2026, 07:27 AM
|
Users browsing this thread: 1 Guest(s)