Recent advances in computational chemistry, artificial intelligence, and structural biology have changed how de novo drug design works. Researchers can now develop new molecules with better pharmacological properties much faster. For example, a newly designed CDK4/6 inhibitor has already reached pre-clinical trials, showing how these technologies are making a real difference in cancer research. This review covers current methods that bring together generative deep learning, molecular modeling, virtual screening, and accurate protein structure prediction in drug discovery. We will focus mainly on generative deep learning models and molecular modeling, with shorter sections on virtual screening and protein structure prediction. This way, readers can easily find the topics most relevant to their interests, generative adversarial networks, reinforcement learning, and diffusion-based methods enable researchers to explore chemical spaces that traditional screening cannot reach. By quantifying the "chemical space unreachable" by classic high-throughput screening (HTS), we can provide a sharper contrast, highlighting the unique capability of these generative models. For example, while traditional HTS might be limited to billions of compounds, generative models can theoretically explore chemical spaces on the order of trillions, offering an advantage of at least three orders of magnitude. Molecular dynamics simulations help predict the stability of ligand-target interactions, identify allosteric sites, and improve docking results. Faster GPUs and better sampling methods have made these advances possible. Hybrid virtual screening methods that mix ligand-based and structure-based approaches often use machine learning to improve results. These techniques help prioritize hits and assess ADMET properties more efficiently. Better protein structure prediction has also expanded the number of druggable targets, especially for proteins without known structures. By combining these computational tools, early drug discovery is faster, less expensive, and produces higher-quality candidates. Together, these changes are reshaping medicinal chemistry.