quant choices for 70B local

[phasestream53]

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

After a few runs on A6000, I ended up with decent performance, but the setup only really made sense once I adjusted around KV cache pressure.

Happy to post more details if anyone wants them.

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

[rocmriley]

Good point. That matches what I saw on dual 3090. The biggest change for me was simplifying the stack instead of adding one more optimization layer.

[benchplease]

Good point. The pattern around batch size has been pretty similar for me.