Large language models (LLMs) have demonstrated remarkable capabilities in various natural language processing tasks, including question answering, text generation, and multimodal understanding. However, their performance in specialized domains such as healthcare and their propensity for generating hallucinated (false) information remains an area of active investigation. This research paper explores the capabilities and limitations of Mistral's LLM, mistral-large-2402, in tackling medical challenge problems and assessing its tendency to hallucinate. The study is motivated by the potential of LLMs to augment medical decision-making processes and the need to evaluate their reliability in critical domains like healthcare. We investigate mistral-large-2402's performance on a curated dataset of medical challenge problems, spanning diagnosis, treatment recommendation, and medical condition analysis tasks. Additionally, we examine the model's propensity for hallucinating by analyzing its responses for factual inconsistencies and unsubstantiated claims. Through quantitative and qualitative analyses, we provide insights into mistral-large-2402's strengths and weaknesses in handling medical challenges. Our evaluation methodology involves measuring the model's accuracy, completeness, and coherence of responses, as well as its ability to recognize and mitigate hallucinations. The findings of this study contribute to the ongoing discourse on the responsible deployment of LLMs in healthcare and highlight potential areas for improvement in model design and training.