Revit Civil AI Estimator

An AI-powered Revit Add-in that automates the Quantity Takeoff and Cost Estimation process for Civil Engineering elements (Walls, Floors, Topography, etc.).

It connects Revit 2025 to OpenAI (GPT-4) to automatically classify elements, assign cost codes (CESMM4/SMM7), and estimate unit rates based on project context.

Features

• Smart Estimation: Automatically extracts geometry and material data from selected elements.
• Context-Aware AI: Allows user to input project context (e.g., "Texas, 2024 Pricing") to adjust estimates.
• Auto-Parameter Creation: Automatically creates Shared Parameters (AI_CostCode, AI_Description, AI_Rate) in your Revit project.
• Async Processing: Runs API calls in the background to prevent Revit UI freezing.

Prerequisites

• Autodesk Revit 2025 (Project targets .NET 8).
• Visual Studio 2022 (with .NET Desktop Development workload).
• OpenAI API Key (You need a paid account to access GPT-4 models).

Installation & Usage

Since this project requires an OpenAI API Key, you must build it yourself.

1. Clone this repository.
2. Open in Visual Studio 2022.
3. Add your API Key:
   • Open the file AiService.cs.
   • Find the line private const string ApiKey = ...
   • Paste your own OpenAI API key (starts with sk-...).
4. Build the Solution:
   • Click Build -> Build Solution.
5. Install to Revit:
   • Go to the output folder bin\Debug\net8.0-windows\.
   • Open RevitAiEstimator.addin and check the path.
   • Copy RevitAiEstimator.addin to %ProgramData%\Autodesk\Revit\Addins\2025\.
6. Run Revit:
   • The "AI Estimator" tab will appear.

Usage

1. Open Revit 2025 and load a model.
2. Go to the "AI Estimator" tab in the Ribbon.
3. Click "Smart Estimate".
4. In the dialog box, enter your Project Context.
   • Example 1: "Austin, Texas. Use local 2024 labor rates. Soil: Limestone."
   • Example 2: "Texas Project. Output in US dollars (USD)."
5. Click Start. The plugin will process elements and populate the properties.
6. Check the Properties Palette of any wall/floor to see the results.

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

License

MIT