Development version 2.55.0 · Stable release used by the beginner guide: 2.49.0 · Free for all of us, forever.
IFCtoLBD turns an IFC building model into Linked Building Data (LBD): small subject–relationship–value facts, called RDF triples, that describe buildings, spaces, elements, and their properties. You can query these facts and use LBD in your own applications to link building information to other data. Development version 2.55.0 supports IFC STEP, IFC/XML, and IFC/JSON.
Audience and prerequisites. The desktop quick start is for BIM practitioners, data engineers, and researchers and requires no programming. Bring an IFC file; the Java distribution also requires Java 21. New to IFC? Read the short introduction below. New to RDF or SPARQL? Start with the conceptual overview, then follow using the triples. Developers need Java 21 and Maven.
Try IFCtoLBD in ten minutes. The checked example bundle includes small IFC models, saved configurations, SPARQL questions, and expected answers for a building inventory, a completeness check, and a revision comparison. The bundle targets development version 2.55.0.
- Convert an IFC model into linked data with the desktop app and save it as Turtle, JSON-LD, or an ICDD package containing the original IFC and separate Linked Building Data submodels.
- Choose what to export: select element types, property sets, and options for properties, units, and geometry.
- Explore your building: query the data with SPARQL, validate it with SHACL, and preview exported geometry in the desktop app.
- Prepare building data for linking: the generated LBD identifiers let you create links to asset records, products, sensors, and other datasets in your own applications. The converter does not create those external links for you.
- Automate your work with the command-line converter, Python examples, or the Java library.
- Build your own tools and extensions from the source: add mappings, vocabularies, validation rules, geometry backends, or domain-specific enrichment.
The guides below show you where to start. The beginner workflow is pinned to release 2.49.0; source-only 2.55.0 features are identified separately.
| Piece | Use it when | Artifact or entry point |
|---|---|---|
| Desktop app | You want the Read IFC → Run interface | IFCtoLBD.zip or IFCtoLBD-Desktop_for_Windows.zip in release 2.49.0 |
| CLI | You want scripted conversion on Windows | IFCtoLBDConverter_CLI.exe in release 2.49.0 |
| Java library | You are embedding conversion in a JVM application | org.linkedbuildingdata:ifc-to-lbd:<version> / ifc-to-lbd-<version>.jar |
| MCP server | You want an AI tool to inspect or convert IFC | IFCtoLBD_MCP; build from source because release 2.49.0 has no prebuilt MCP server |
These are the exact names for release 2.49.0. New releases follow the artifact naming policy.
Industry Foundation Classes (IFC) is an open, vendor-neutral standard from buildingSMART for exchanging building and infrastructure information. An IFC model can describe geometry, elements, spaces, properties, and relationships. Read buildingSMART’s introduction to IFC if you need the background. For technical definitions, follow its links to the IFC specifications. You can skip this introduction if you already use IFC.
You need an IFC model and the desktop application.
-
Open the 2.49.0 release, expand Assets, and choose
IFCtoLBD.zipfor the Java 21 distribution orIFCtoLBD-Desktop_for_Windows.zipfor the bundled Windows application. The bundled Windows application includes Java. The Java distribution needs Java 21.
-
Start the desktop application. For a result you can check, use the small two-wall example model. Click Read IFC, choose the model, and wait for it to finish reading. Development version 2.55.0 also shows a model summary with the detected schema, model contents, geometry availability, and warnings. The application proposes an output path.
-
Check that output path, then click Run. For your first conversion, keep the existing settings and filters. Wait for the conversion log to report completion.
-
In development version 2.55.0, open Query, paste the saved building inventory query, and click Run query. The useful result is immediate: the storey contains two walls. Compare it with the expected result.
Follow the release-pinned desktop quick start for launch commands, output choices, and help if the application does not start.
Windows command-line alternative: IFCtoLBDConverter_CLI.exe is the Windows
command-line converter. Use it from PowerShell or Command Prompt;
see the Windows CLI instructions.
The same functionality is available as a Java app for Linux and macOS users.
| Your goal | Guide |
|---|---|
| Understand LBD and use the generated triples | Conceptual overview, then your first queries |
| Browse the generated LBD as an interactive graph | LBD Browser |
| Read the output or automate conversion in Python | Python guide |
| Embed the converter and query its output in Java | Java guide |
| Use AI to inspect, query, validate, and convert IFC models | MCP tutorial — build from source; release 2.49.0 has no prebuilt MCP package |
| Compile the converter, understand the subprojects, or extend it | Build and development guide |
| Explore classification, product identity, GS1, or EPD output | Supply-chain and sustainability |
Start with the output you have just created; programming and compiling can come later. LBD uses shared vocabularies such as the Building Topology Ontology (BOT). The W3C Linked Building Data Community Group provides further background and community resources.
Browse the documentation website or the documentation index. Earlier announcements are in project history.
The citation authors, in the same order as CITATION.cff, are Jyrki Oraskari,
Mathias Bonduel, Kris McGlinn, Pieter Pauwels, Freddy Priyatna, Anna Wagner,
Ville Kukkonen, Simon Steyskaland, Joel Lehtonen, and Maxime Lefrançois.
Additional contributions were made by Lewis John McGibbney. Thanks also to
Vladimir Alexiev, Kathrin Dentler, and Lukas Kirner for their valuable insights.
This project is released under the open source Apache License, Version 2.0
@misc{jyrki\_oraskari\_2026\_07,
author = {Jyrki Oraskari and
Mathias Bonduel and
Kris McGlinn and
Pieter Pauwels and
Freddy Priyatna and
Anna Wagner and
Ville Kukkonen and
Simon Steyskaland and
Joel Lehtonen and
Maxime Lefrançois},
title = {{IFCtoLBD v 2.55.0}},
month = 07,
year = 2026,
publisher = {GitHub},
version = {2.55.0},
url = {https://github.com/jyrkioraskari/IFCtoLBD}
}
The research was partly funded by the EU through the H2020 project BIM4REN.