+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Hardware (https://www.datasculpting.ai/inference/forumdisplay.php?fid=4)
+--- Forum: CPUs / NPUs / Accelerators (https://www.datasculpting.ai/inference/forumdisplay.php?fid=30)
+--- Thread: consumer accelerators worth watching with Ollama (/showthread.php?tid=1178)
Pages:
1
2
consumer accelerators worth watching with Ollama - mangofield33 - 10-13-2026
Did a bit of testing and thought I'd post the result here.
After a few runs on A6000, I ended up with decent performance, but the setup only really made sense once I adjusted around tool calling.
Happy to post more details if anyone wants them.
For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.
If I had to summarize the lesson so far, it would be that clean setup and observability beat clever guesses.
Re: consumer accelerators worth watching with Ollama - vramridge - 10-14-2026
Good point. That matches what I saw on M3 Max.
Re: consumer accelerators worth watching with Ollama - bytesupply76 - 10-15-2026
Quote:Good point. That matches what I saw on M3 Max.
I'd approach it in a boring order. If you're on MLX, I'd confirm the version first and then reduce variables until the behavior is obvious. Once you know the plain path works, then add the faster path back in piece by piece.
Re: consumer accelerators worth watching with Ollama - phaseport36 - 10-15-2026
Quote:Quote:Good point. That matches what I saw on M3 Max.
Tried to reproduce it just now. The issue went away once I changed one setting and restarted the stack cleanly. The surprise was that memory pressure showed up before throughput collapsed.
If you post your exact settings, people here can probably narrow it down pretty fast.
Re: consumer accelerators worth watching with Ollama - rocmmixer - 10-15-2026
A couple things I'd check first. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at reranking. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
Re: consumer accelerators worth watching with Ollama - phasetrack - 10-18-2026
I hit almost the same issue last night.
Re: consumer accelerators worth watching with Ollama - sanescope - 10-18-2026
Interesting, I hadn't tried that. I ended up in about the same place using TensorRT-LLM. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.
Re: consumer accelerators worth watching with Ollama - marbletrack - 10-19-2026
What worked for me was pretty simple. On Threadripper box, the easiest win is often to back off the aggressive settings and see whether the problem survives. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
Re: consumer accelerators worth watching with Ollama - homelabhenry - 10-19-2026
What worked for me was pretty simple. On dual 3090, the easiest win is often to back off the aggressive settings and see whether the problem survives. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
Re: consumer accelerators worth watching with Ollama - rocmriley - 10-19-2026
Ran a quick check on my box. My result was mixed: first run looked bad, second run looked normal. The main thing that moved was latency, not quality.