mradermacher/sokol-1.5b-02-i1-GGUF
Quantization vs kristaller486/sokol-1.5b-02. mradermacher/sokol-1.5b-02-i1-GGUF is a GGUF quantization (Q6_K) of the same weights: 133 tensors match exactly and 464 within quantization error.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/133, MLP 0/198, Attention 0/264