kairusama/Qwen3.6-27B-Omnimerge-v4-MTP-IQ3_XXS-GGUF
Quantization vs ManniX-ITA/Qwen3.6-27B-Omnimerge-v4-MTP-GGUF. kairusama/Qwen3.6-27B-Omnimerge-v4-MTP-IQ3_XXS-GGUF is a GGUF quantization (IQ3_XXS) of the same weights: 360 tensors match exactly and 506 within quantization error.
Tensors changed: lm_head 0/1, other 0/290, Norms 0/216, Attention 0/164, MLP 0/195