mradermacher/Umbra-v3-MoE-4x11b-GGUF
Quantization vs SteelStorage/Umbra-v3-MoE-4x11b. mradermacher/Umbra-v3-MoE-4x11b-GGUF is a GGUF quantization (Q8_0) of the same weights: 145 tensors match exactly and 194 within quantization error.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/97, Attention 0/192, other 0/48