IFC-Copilot framework for AI-assisted building design accepted at EG-ICE 2026

A paper by the research group of Prof. Christian Bartelt, in cooperation with the research group of Prof. Stefan Lüdtke at the University of Rostock, has been accepted for presentation at the 33rd EG-ICE International Workshop on Intelligent Computing in Engineering.

The paper, titled “IFC-Copilot: A Tool-Based Framework for LLM-Driven IFC Building Design,” presents an open-source framework that enables Large Language Models to interact directly with digital building models. The work focuses on Industry Foundation Classes, or IFC, an open standard for representing building information in Architecture, Engineering, and Construction workflows.

Bringing generative AI into engineering practice requires more than producing text or design suggestions. AI systems must be able to translate natural language instructions into reliable actions on structured engineering data. IFC-Copilot addresses this challenge by allowing language models to query, create, and edit IFC-based Building Information Modeling, or BIM, models through a tool-based interface.

The framework combines predefined BIM tools, dynamic IFC-specific code generation, and retrieval-augmented generation based on IFC documentation. Through the Model Context Protocol, IFC-Copilot provides a standardized interface that allows language models to work with BIM models independently of proprietary BIM software APIs. This makes the approach flexible, transparent, and easier to evaluate across different workflows.

In evaluations with Claude Sonnet 4.5 and GPT-5.4, IFC-Copilot achieved strong performance on custom IFC benchmarks and IFC-bench-v1. Interactive multi-turn examples further show how AI systems can support BIM design processes by carrying out design-related changes over several steps while working directly with standardized building model data.

The work closely reflects the vision behind the new AI Engineering bachelor’s program at TU Clausthal: combining artificial intelligence with engineering expertise to develop reliable, practical, and critically evaluated AI-supported systems for real-world applications.

More information about the AI Engineering bachelor’s program is available here.

The paper website is available here.