vinhnx90/gemma-3-1b-thinking-v2-Q4_K_M-GGUF
Quantization vs vinhnx90/gemma-3-1b-thinking-v2. vinhnx90/gemma-3-1b-thinking-v2-Q4_K_M-GGUF is a GGUF quantization (Q4_K_M) of the same weights: 157 tensors match exactly and 183 within quantization error.
Tensors changed: Norms 0/157, Embeddings 0/1, MLP 0/78, Attention 0/104