Posts: 184
Threads: 23
Joined: Sep 2026
Reputation:
0
Thought I'd post a small recipe because a lot of threads ask similar things.
This isn't revolutionary, but the combination of Ollama and 7900 XTX felt surprisingly usable once I tuned around tool calling.
If someone wants the exact settings I can clean them up and post them.
Mostly CUDA, sometimes wrong, usually measured
Posts: 46
Threads: 7
Joined: Sep 2026
Reputation:
0
I don't think that's the whole story. I get why that seems right, but on mini PC the bottleneck felt elsewhere. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.
There are a few different conversations hiding inside this thread, so being explicit about the goal helps.
Posts: 41
Threads: 4
Joined: Sep 2026
Reputation:
0
Tried to reproduce it just now. On dual 3090 with SGLang, I saw basically the same behavior.
Posts: 25
Threads: 2
Joined: Sep 2026
Reputation:
0
What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at consumer GPUs. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
Posts: 32
Threads: 2
Joined: Sep 2026
Reputation:
0
Good point. That matches what I saw on 4090. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.
Posts: 41
Threads: 0
Joined: Oct 2026
Reputation:
0
That matches what I was seeing. The pattern around tokens/sec has been pretty similar for me.
