I got Mimo 2.5 Pro running on my 8x Asus Nvidia GB10 cluster using mtp-2, single user request, coding: 40 t/s - 1k context, 32t/s - 30k context, 25t/s - 125k context, 17t/s - 250k context. 2 parallel reached 60t/s and in 4 parallel reached 83t/s, not bad for 1T model. Works just fine with open code
