keisuke-miyako/multilingual-e5-small-gguf-q4_k_m
Quantization vs intfloat/multilingual-e5-small. keisuke-miyako/multilingual-e5-small-gguf-q4_k_m is a GGUF quantization (Q4_K_M) of the same weights: 124 tensors match exactly and 73 within quantization error.
Tensors changed: Embeddings 0/5, Norms 0/48, Attention 0/96, other 0/48