CobrIX/CobrIX-1.0-Coder-Full-72B-A18B-GGUF
Not comparable yet vs empero-ai/Qwythos-9B-v2. Not verifiable yet: the GGUF's tensors don't map onto the other model's names (283 of 775/587 match), usually an architecture llama.cpp's name map doesn't cover yet.