Currently, the use of Large Language Models (LLMs) has spread to different areas of knowledge, and new versions are constantly being released that offer new information updates. However, it is necessary to study them to understand their scope. In this article, an LLM is explored to determine if it can interpret a truss diagram and, through prompt engineering, evaluate its limitations. The results found that the LLM used identifies the elements of a truss from an image and can assist the user in calculating axial forces. Finally, recommendations are shared for using LLMs to perform calculations with trusses.