mistrjirka/Qwen3.8-27B-PXQ-GGUF
Quantization vs Qwen/Qwen3.8-27B. mistrjirka/Qwen3.8-27B-PXQ-GGUF is a GGUF quantization (T254/Q8_0) of the same weights: 161 tensors match exactly and 34 within quantization error.
Tensors changed: Norms 0/161, Attention 0/32, Embeddings 0/1, lm_head 0/1