A new initiative is underway to enhance the application of large language models (LLMs) in biomedical research. The platform, named BioChatter, aims to address the limitations of commercial platforms by providing an open-source solution that emphasizes transparency and customization. This effort is part of a broader evaluation of computational accuracy in numerical reasoning tasks for healthcare applications. Additionally, discussions are ongoing regarding the adequacy of current methods for evaluating medical AI systems to ensure their clinical reliability. Recent studies, including one published in npj Digital Medicine, are focused on building versatile large language models specifically for medical use.