mradermacher/Beyonder-4x7B-v3-GGUF
Quantization vs mlabonne/Beyonder-4x7B-v3. mradermacher/Beyonder-4x7B-v3-GGUF is a GGUF quantization (Q8_0) of the same weights: 97 tensors match exactly and 130 within quantization error.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/65, Attention 0/128, other 0/32