EPlus-LLM

EPlus-LLM series, natural language for auto-building energy modeling via LLM 🎉 News - ⚠️ [2025/05/15] (update #5): A bug has been fixed and the model has been updated. Many thanks to the user for the feedback! Appreciated!! - 📄 [2025/04/18] (update #4): The paper related to the EPlus-LLMv2 platform has been accepted for publication in Automation in Construction. Paper here. - ⚡️ [2025/01/15] (update #3): We release EPlus-LLMv2, successfully addressing the challenge of auto-building energy modeling (ABEM) in complex scenarios. The new version of the platform supports a wide range of modeling scenarios encountered in real-world building applications, significantly enhancing its breadth and flexibility. Based on comprehensive datasets and a large-scale LLM, we integrate techniques such as LoRA, mixed precision training, and model quantification to reduce computational burden and achieve efficient fine-tuning (without compensating performance). Paper coming soon. - 📄 [2025/01/14] (update #2): Our paper on using prompt engineering to inform LLMs for automated building energy modeling has been accepted by Energy. Paper here. - 🔥 [2024/05/016] (update #1): We first successfully implement natural language-based auto-building modeling by fine-tuning a large language model (LLM). Paper here. 🚀 Key Features - Scalability: Auto-generates complex EnergyPlus models, including varying geometries, materials, thermal zones, hourly schedules, and more. - Accuracy & Efficiency: Achieves 100% modeling accuracy while reducing manual modeling time by over 98%. - Interaction & Automation: A user-friendly human-AI interface for seamless model creation and customization. A user-friendly human-AI interface for EPlus-LLMv2. ✅ Geometry: square, L-, T-, U-, and hollow-square-shaped buildings ✅ Roof types: flat, gable, hip – customizable attic/ridge height ✅ Orientation & windows: custom WWR, window placement, facade-specific controls ✅ Walls & materials: thermal properties, insulation types ✅ Internal loads: lighting, equipment, occupancy, infiltration/ventilation, schedules, heating/cooling setpoints ✅ Thermal zoning: configurable multi-zone layouts with core & perimeter zones 🏗️ Target Users This current platform is designed for engineers, architects, and researchers working in building performance, sustainability, and resilience. It is especially useful during early-stage conceptual design when modeling decisions have the greatest impact. Here provides a code snippet to show you how to load the EPlus-LLM and auto-generate building energy models. [](https://colab.research.google.com/github/Gangjiang1/EPlus-LLM/blob/main/v2/EPlus-LLMv2inference.ipynb) If you find our work helpful, feel free to give us a cite.