Pie & AI:Vienna – Maximizing Inferencing Efficiency with Model Quantization