+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Evaluation & Testing (https://www.datasculpting.ai/inference/forumdisplay.php?fid=16)
+--- Forum: Benchmarks (https://www.datasculpting.ai/inference/forumdisplay.php?fid=65)
+--- Thread: benchmark result I don't trust: early impressions (/showthread.php?tid=1264)
Pages:
1
2
benchmark result I don't trust: early impressions - kernelarch - 10-15-2026
I'm trying to sanity-check my setup.
I'm mostly working with DeepSeek in vLLM on GB200. The part I'm trying to understand is tokens/sec.
Not looking for the perfect answer, mostly what worked in real use.
Re: benchmark result I don't trust: early impressions - plainline - 10-16-2026
My first guess would be this: On mini PC, the easiest win is often to back off the aggressive settings and see whether the problem survives. Once you know the plain path works, then add the faster path back in piece by piece.
Re: benchmark result I don't trust: early impressions - batchpeak - 10-16-2026
worth checking the driver first.
Re: benchmark result I don't trust: early impressions - batchsize - 10-16-2026
subscribed, curious where this lands.
If you post your exact settings, people here can probably narrow it down pretty fast.
Re: benchmark result I don't trust: early impressions - cacheline - 10-17-2026
My first guess would be this: On A6000, 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: benchmark result I don't trust: early impressions - routearch - 10-17-2026
That matches what I was seeing. I don't think your conclusion is far off. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
Re: benchmark result I don't trust: early impressions - batchsize - 10-18-2026
Need a little more detail here. Are you testing this on A6000 or something else?
Re: benchmark result I don't trust: early impressions - deltastation - 10-18-2026
I'd approach it in a boring order. On B200, the easiest win is often to back off the aggressive settings and see whether the problem survives.
Re: benchmark result I don't trust: early impressions - kernelghost - 10-18-2026
Quote:I'd approach it in a boring order. On B200, the easiest win is often to back off the aggressive settings and see whether the problem survives.
Maybe, but I had the opposite result. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.
Re: benchmark result I don't trust: early impressions - cacheline - 10-18-2026
A couple things I'd check first. If you're on Ollama, I'd confirm the version first and then reduce variables until the behavior is obvious. Not glamorous advice, but repeatable baselines save a lot of time.