Objective This article takes the construction of the Huaxi Hongyi Medical University model as the core, explores its application effect in assisting the generation of medical records, and provides a practical path for the construction and application of artificial intelligence in medical institutions. Methods Through strategies such as multimodal data fusion, domain adaptation training, and localization of hardware adaptation, a large-scale medical model with 72 billion parameters was constructed. Combined with technologies such as speech recognition, knowledge graphs, and reinforcement learning, an application system for assisting in the generation of medical records was developed. Results Taking the assisted generation of discharge records as an example, in the pilot department, the average completion time of writing was reduced from 21 minutes to 5 minutes (a decrease of 76.2%), the accuracy rate of the model output reached 92.4%, and the annotation consistency Kappa coefficient was 0.85. Conclusion The model of medical institutions constructing independently controllable large-scale medical models and incubating various applications based on them can provide a reference path for the artificial intelligence construction of similar institutions.