andrew-paul/Qwen3.8-27B-DFlash2-Q3_K_M-GGUF
Quantization vs incoai/Qwen3.8-27B-DFlash2. andrew-paul/Qwen3.8-27B-DFlash2-Q3_K_M-GGUF is a GGUF quantization (Q3_K_M) of the same weights: 21 tensors match exactly and 36 within quantization error.
Tensors changed: other 0/1, Norms 0/21, MLP 0/15, Attention 0/20