containment for model-driven automation: worth it?

[bytesupply76]

subscribed, curious where this lands.

[vramarch]

I tested this tonight. The issue went away once I changed one setting and restarted the stack cleanly. The main thing that moved was latency, not quality.

[dockkite69]

Thanks, that helps. 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.