mmnga/llm-jp-3-980m-instruct3-gguf
Quantization vs llm-jp/llm-jp-3-980m-instruct3. mmnga/llm-jp-3-980m-instruct3-gguf is a GGUF quantization (Q8_0) of the same weights: 41 tensors match exactly and 142 within quantization error.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/41, MLP 0/60, Attention 0/80