Quantize Meta-Llama-3-8B-Instruct with llamafile-0.7.1 Q8_0 48dfbf8 verified jartine commited on Apr 19