<p>Gasification is an important procedure because it efficiently transforms a variety of carbon-containing substances like coal, biomass, and waste into valuable products such as syngas, hydrogen, and synthetic fuels. The effectiveness of the gasification procedure relies on important factors that need to be taken into account. Thus, in the present study, a machine learning (ML) technique was used based on initial data from 312 experiments in order to predict the production rate of raw gases and solids in the gasification process. Over the modeling stage, as the novelty of the research, metaheuristic optimization algorithms as successors were coupled to the Decision Tree (DT) as the main model to tune the performance of the model, leading to simulate gasification efficiently compared to the single model. Raw gases and solids generated at the end of gasification depend on various factors, including input materials and thermodynamic parameters, e.g., temperature. By applying the developed hybrid models with optimizers COOT optimization algorithm (COA) and Gradient-based optimizer (GBO), it becomes feasible to create a framework quantifying the influence of the variables on the production of gases and solids. Several assessing criteria were employed to evaluate the performance of developed models. By investigating the results of modeling, hybrid models were found at a higher level of modeling in terms of coefficient of determination with the average calculated R² of around 98%, while for a single model of 96%. Generally, the developing approach of the bare single AI-based models can be used to simulate thermodynamic operations to give experts more understanding of processes working efficiently.</p>