+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Miscellaneous Discussion (https://www.datasculpting.ai/inference/forumdisplay.php?fid=24)
+--- Forum: General (https://www.datasculpting.ai/inference/forumdisplay.php?fid=25)
+--- Thread: new here, what's your stack with SGLang (/showthread.php?tid=77)
Pages:
1
2
new here, what's your stack with SGLang - thetagrid15 - 08-31-2026
Thought I'd post a small recipe because a lot of threads ask similar things.
Just sharing a setup that ended up working reasonably well: DeepSeek, PyTorch, mini PC, with a lot of attention on speculative decoding.
If someone wants the exact settings I can clean them up and post them.
For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.
Re: new here, what's your stack with SGLang - thetagrid15 - 08-31-2026
that fixed it for me too.
Re: new here, what's your stack with SGLang - tensorkite - 09-01-2026
What exact setup are you running? Are you testing this on 7900 XTX or something else?
Re: new here, what's your stack with SGLang - ggufwire22 - 09-01-2026
that fixed it for me too.
Re: new here, what's your stack with SGLang - syshouse - 09-02-2026
Need a little more detail here. Do you see the same thing with a smaller batch or shorter prompt?
Re: new here, what's your stack with SGLang - griffinloom66 - 09-05-2026
That matches what I was seeing. The pattern around tokens/sec has been pretty similar for me. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.
Re: new here, what's your stack with SGLang - dockember - 09-08-2026
Good point. I don't think your conclusion is far off.
Re: new here, what's your stack with SGLang - plaingarden18 - 09-08-2026
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.
Re: new here, what's your stack with SGLang - quiettoken - 09-08-2026
A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with Phi and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.
Re: new here, what's your stack with SGLang - rocmplane - 09-09-2026
I tested this tonight. 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.