jashepp/Ornith-1.0-9B-MXFP4_Hybrid-Imatrix-GGUF
Quantization vs ornith-ai/Ornith-1.0-9B. jashepp/Ornith-1.0-9B-MXFP4_Hybrid-Imatrix-GGUF is a GGUF quantization (Q8_0/F16) of the same weights: 83 tensors match exactly and 128 within quantization error.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/81, MLP 0/96, Attention 0/32