The health, comfort, and well-being of both humans and animals are directly affected by the air quality index (AQI), making its monitoring essential for multiple reasons. By predicting toxic gases and AQI matter quickly, allowing for immediate actions to enhance air quality. Some machine learning models exist that operate in solo mode, predicting the AQI index without yielding a satisfactory result. Therefore, it is necessary to hybridize neural networks to accurately predict the air quality prediction index, enabling the concerned departments to take appropriate action. As a result, the proposed model considers not only air quality index data, but also weather data, which is a blend of an artificial neural network and a bidirectional Convolution Neural network ( CNN)-Long short term neural network(LSTM) model. The genetic algorithm catalyzes this hybrid model to accurately predict the air quality index. The rigorous evaluation of the implemented model for the parameters RMSE and accuracy yields good results of 2.987 and 98.98, respectively, which is again far better than the other solo models that predict the air quality index.