Qwen3.8-Flash-Next (UD-IQ4_XS) on 2x RTX 3060 + 7800X3D, from initial 36 tps prefill to 400 tps and other benchmarks (-sm tensor trap) + VRAM/RAM usage

TL;DR: llama.cpp with --load-mode mmap used 21-32 GB of RAM, with ik_llama.cpp using 106-108 GB. -sm tensor killed my prefill, changing to -sm layer went from 36 tps to 135 tps. After that, -ubatch 2048 pushed it up to 400 tps at -c 131072. I can't fit -ubatch 2048 at -c 262144, which is why it caps