mradermacher/Mistral-Large-3-675B-Base-2512-i1-GGUF
Quantization vs mistralai/Mistral-Large-3-675B-Base-2512. mradermacher/Mistral-Large-3-675B-Base-2512-i1-GGUF is a GGUF quantization (IQ2_XS) of the same weights: 181 tensors match exactly and 357 within quantization error.
Tensors changed: Norms 0/123, Attention 0/183, other 0/58, MLP 0/174