bartowski/deepcogito_cogito-v2-preview-llama-109B-MoE-GGUF
Quantization vs deepcogito/cogito-v2-preview-llama-109B-MoE. bartowski/deepcogito_cogito-v2-preview-llama-109B-MoE-GGUF is a GGUF quantization (Q8_0) of the same weights: 145 tensors match exactly and 338 within quantization error.
Tensors changed: lm_head 0/1, Embeddings 0/1, MLP 0/192, Norms 0/97, Attention 0/192