

Last night I managed to run Qwen Flash Next iq2 XS using Strata in docker on a VM
I passed through my 3090, 64GB of DDR4, 12 vcores of my 5960x,
Managed 90t/s
I’ll probably try a larger version soon but QFN already seems a lot more capable than 3.6 35b A3B, and runs better than Qwen 3.8 dense in llama.cpp



One of the few silver linings of the rampocalypse that I’ve noticed is a small increase in programs being developed to use less memory.