The implementation and rapid development of Artificial Intelligence have greatly contributed to the advancement of natural sciences, such as Chemistry. Advancements in Machine Learning and deep learning have revolutionized chemical research, reducing reliance on time-consuming and costly laboratory experiments. These technologies enable faster and more accurate predictions of reaction outcomes, accelerating innovation by analyzing vast datasets of chemical reactions, identifying patterns, and optimizing synthesis routes. This efficiency boost is particularly impactful in drug development, reducing the time to market for new medications. The integration of multi-agent systems with Large Language Models (LLMs) and molecular transformers further enhances AI's role in Chemistry. These agents facilitate advanced decision-making, task execution, and problem-solving by interacting with external tools, databases, and experimental platforms. Prompt Chaining, Molecular Transformers, and Retrieval-Augmented Generation (RAG) enhance Large Language Models (LLMs) in chemistry and drug discovery by enabling step-by-step problem-solving, accurate reaction prediction, and informed decision-making through real-time data retrieval, significantly improving AI-driven research and molecular design. By leveraging LLMs for natural language understanding and reasoning, multi-agent systems improve efficiency, sustainability, and safety in chemical synthesis, as demonstrated in aspirin synthesis.