Autism spectrum disorder (ASD) is a complex condition that affects who the brain develops and functions, due to its heterogeneous nature often challenges traditional symptom-based diagnostic techniques. Advancements in non-invasive, radiological techniques such as resting-state functional magnetic resonance imaging (rs-fMRI) offers researchers to work on brain connectivity patterns that can help in diagnosis of ASD. Early diagnosis in case of ASD is very important for early intervention and personalized treatment. This study utilizes the pre-processed RS-fMRI data from the ABIDE—I database and computational intelligence to enhance the diagnosis of ASD. The study proposes a novel Hybrid model called the Stacked Ensemble model, integrating Support Vector machines (SVM), Gradient Boosting (XGBoost), and Random Forest (RF) as base learners and a Logistic regression classifier as the final estimator. The stacking ensemble achieved an accuracy of ~ 0.71. This approach demonstrates superior performance compared to standalone classifiers and highlights the potential of combining RS-fMRI data with Machine Learning to improve ASD diagnosis.