Baekpica/Inkling-Small-GGUF
Quantization vs thinkingmachines/Inkling-Small. Baekpica/Inkling-Small-GGUF is a GGUF quantization (Q8_0) of the same weights: 722 tensors match exactly and 166 within quantization error.
Tensors changed: MLP 0/328, Attention 0/378, Norms 0/174, other 0/5, Embeddings 0/3