QQZ2026/Qwen3.8-27B-ZeroRefusal-UD-IQ4_XS-MTP-GGUF
Not comparable yet vs Qwen/Qwen3.8-27B. Not verifiable yet: the GGUF's tensors don't map onto the other model's names (0 of 1199/992 match), usually an architecture llama.cpp's name map doesn't cover yet.