working config for 32k context

[northgrid33]

This keeps coming up in conversation so I figured I'd start a thread.

It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with long context.

Maybe obvious to some of you, but it clicked for me this week.

[vramstream]

I'd approach it in a boring order. On 5090, the easiest win is often to back off the aggressive settings and see whether the problem survives.