EURAC Research and BAEDA Lab release the Python library BrickLLM

The rapid rise of (LLMs) is opening new doors for building data management and high of data-driven energy management solutions in buildings.
Imagine a future where creating metadata models for buildings is no longer a tedious, manual task. With the help of LLMs, we can automate template generation based on detailed descriptions of building components and equipment provided by professionals—making the process faster, more efficient, and accessible to a broader audience.

In this framework, the Energy Efficient Building group (at Eurac Research), particularly through the efforts of Marco Perini, Daniele Antonucci, Olga Somova and the support of Cristian Pozza, together with our BAEDA Lab members  Giudice Rocco, Piscitelli Marco Savino and  Capozzoli Alfonso have released brickllm, the Python library which allows for the generation of semantic metadata models of the building using Brick ontology along with its related application. Pypi: click here.

BrickLLM is a Python library for generating RDF files following the BrickSchema ontology using LLMs.

Main features of BrickLLM:

• Generate BrickSchema-compliant RDF files from natural language descriptions of buildings and facilities
• Support for multiple LLM providers (OpenAI, Anthropic, Fireworks)
• Customizable graph execution with LangGraph
• Easy-to-use API for integrating with existing projects