Application of Deep Generative Models in the Design of Buildings and Structures

This article explores the possibilities of using generative models in the design of capital construction facilities and their interiors. The article presents practical and theoretical research methods for testing several models, including diffusion generative models and generative transformers-based language models. The results are divided into three groups: working with images, working with text and writing programs. The article shows that generative models can simplify and speed up the work of designers. However, the accuracy of the models needs to be improved. In general, the results indicate that generative models can be effectively used for tasks that do not require high accuracy.