cstr/smt-grandstaff-GGUF
Quantization vs antoniorv6/smt-grandstaff. cstr/smt-grandstaff-GGUF is a GGUF quantization (Q8_0) of the same weights: 230 tensors match exactly and 111 within quantization error.
Tensors changed: Attention 0/128, MLP 0/32, Norms 0/80, Embeddings 0/4, other 0/97