new here, what's your stack with SGLang

[thetagrid15]

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.

[thetagrid15]

that fixed it for me too.

[tensorkite]

What exact setup are you running? Are you testing this on 7900 XTX or something else?

[ggufwire22]

that fixed it for me too.

[syshouse]

Need a little more detail here. Do you see the same thing with a smaller batch or shorter prompt?

[griffinloom66]

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.

[dockember]

Good point. I don't think your conclusion is far off.

[plaingarden18]

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.

[quiettoken]

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.

[rocmplane]

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.