Plant breeding is crucial for improving crops with desirable traits, and phenotypic prediction plays a key role in this process. Accurately predicting plan height is a challenge but is vital for optimizing crop management, maximizing yield, improving disease resistance, enhancing environmental adaptation, ensuring efficient harvesting, and streamlining breeding programs. Our analysis utilized GWAS and implemented machine learning (ML) and deep learning (DL) models. We achieved more than 75% accuracy in classifying rice subpopulation and an accuracy range of 0.64–0.76 for predicting rice plant height based on genotype information. Using these models will enhance productivity and sustainability of agriculture. With this methodology and results, one can reasonably estimate that deep learning models are best for predicting large data and machine learning models are reasonably good for small set of data.