Ollama for power users

[mellowwire98]

Did a bit of testing and thought I'd post the result here.

After a few runs on GB200, I ended up with decent performance, but the setup only really made sense once I adjusted around structured output.

Happy to post more details if anyone wants them.

[dockridge]

I'd start with the simpler stack.

[batchscope72]

Tried to reproduce it just now. I couldn't reproduce the slowdown until I pushed context much higher.