The main aim of this work is to propose a hybrid framework, which makes use of intelligent decision-making tools, that are gray, fuzzy, and ANFIS, to optimize the multi-performance characteristics (MPCs) of powder-mixed electrical discharge machining (PM-EDM) of tungsten carbide (WC). To perform the experimentation, four input parameters: (i) pulse-on time, (ii) current, (iii) powder concentration, and (iv) powder grain size are considered to investigate the MPCs such as material removal rate, tool wear rate, surface roughness, and micro-hardness. The proposed framework uses response surface methodology (RSM) with gray, gray-fuzzy, and gray-ANFIS approaches to obtain optimal solution and also to handle the element of uncertainty or fuzziness associated with the uncertain, multi-input, and discrete data. This method helps to generate the values of gray relational grade (GRG), gray-fuzzy reasoning grade (GFRG), and gray adaptive neuro-fuzzy inference system grade (G-ANFISG) for all the 30 experiments. Analysis of variance (ANOVA) is performed on GRG, GFRG, and G-ANFISG to identify the major contributing input parameters which may affect the MPCs. Finally, the theoretical prediction is done to verify the improvement in the performance characteristics obtained through the proposed approaches. Both the experimental and statistical results clearly demonstrate the success of proposed framework for the optimization of PM-EDM of WC alloy.