allura-org/MoE-Girl-1BA-7BT-GGUF
Quantization vs allura-org/MoE-Girl-1BA-7BT. allura-org/MoE-Girl-1BA-7BT-GGUF is a GGUF quantization (Q8_0) of the same weights: 81 tensors match exactly and 66 within quantization error.
Tensors changed: MLP 0/16, Embeddings 0/1, Norms 0/65, Attention 0/64, lm_head 0/1