disaggregated inference thoughts with llama.cpp

[sparrowkite53]

I'd start with the simpler stack.

[rocmhouse90]

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. Could be a workload difference, but I wouldn't generalize too fast from one setup.

[stackarch]

Did a small comparison this afternoon. On 7900 XTX with PyTorch, I saw basically the same behavior. I expected the model choice to matter more than it did.

[kernelrunner]

Did a small comparison this afternoon. On 7900 XTX with PyTorch, I saw basically the same behavior. I expected the model choice to matter more than it

Appreciate the detail. I ended up in about the same place using llama.cpp. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

There are a few different conversations hiding inside this thread, so being explicit about the goal helps.