containment for model-driven automation

[cudapoint]

Thought I'd post a small recipe because a lot of threads ask similar things.

Maybe useful as a starting point: llama.cpp on Threadripper box, keeping MoE serving in mind from the start.

If someone wants the exact settings I can clean them up and post them.

I care more about stability than squeezing the last possible tokens/sec out of it.

[rocmriley]

Thought I'd post a small recipe because a lot of threads ask similar things.

Interesting, I hadn't tried that. The pattern around embeddings has been pretty similar for me. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[phaseshed]

I'd start with the simpler stack.

[thetafield]

Maybe, but I had the opposite result. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Could be a workload difference, but I wouldn't generalize too fast from one setup.

[rocmstack25]

A couple things I'd check first. On dual 3090, the easiest win is often to back off the aggressive settings and see whether the problem survives.

[thetaarch43]

What worked for me was pretty simple. If you're on llama.cpp, I'd confirm the version first and then reduce variables until the behavior is obvious.

[mangopeak72]

What worked for me was pretty simple. If you're on llama.cpp, I'd confirm the version first and then reduce variables until the behavior is obvious.

A couple things I'd check first. If you're on LM Studio, I'd confirm the version first and then reduce variables until the behavior is obvious.

[torchdesk]

Need a little more detail here. Is this local-only or are you routing anything externally?

[cobaltspark]

A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there. Once you know the plain path works, then add the faster path back in piece by piece.

[dockember]

following this because I'm in the same boat.