AI quantity takeoff from IFC models
Ask your AI for wall areas, concrete volumes or beam lengths, per element, material or storey, and moddl pulls the numbers from your IFC model. You see whether a value comes from the file or was measured, and how many elements went into a sum.
01Which quantities your AI takes off
Ask the way you’d ask a colleague in cost estimating, and your AI turns it into calls to moddl. A few examples:
What volume do the walls have, per storey?
Sum up the wall areas per material.
How many metres of beams does the model hold, and what do they weigh?
Measure the volume of the footings, also where the file holds no quantities.
02Where every number comes from
For every quantity you see where it comes from. A value from the IFC file is marked file. If moddl measured it on the element’s geometry, you see measured and the method. If it rests on a named rule, it’s an assumption, and so is any sum with an assumed value in it.
If a quantity can’t be measured, the answer gives you the reason rather than a 0. moddl also normalises units, so if the file is in millimetres, you still get metres.
Your AI can also ask for both. Where the file value and the measured one disagree, the file’s quantities don’t match the geometry.
Two tools sit behind this. Your AI adds up values from the file with find_elements, split into groups if you like, and compute_quantities measures volume, areas, length and mass on the geometry.
03How complete a sum is
A sum with six walls missing looks just like a complete one. That’s why moddl gives each result its coverage: how many of the requested elements have a value or could be measured, say 40 of 46 walls. For the other six you learn what went wrong.
You get the reason as a code, for instance no_plane_pair when an element has no pair of faces for a side area, or density_not_declared when a mass is missing its density.
04What an answer looks like
Each quantity comes back as a number with its unit, together with where it comes from and how many elements it counts, per element or as a sum. Where there’s no number, you get the reason and the next step, for example: enter the density with its source, then measure again.
If you want a quantity split by material or storey, your AI groups it with group_by, and each group gets its own sum with its own count of elements.
Elements without a storey are counted separately rather than quietly added to one. Typical cases are parts of an assembly with no storey of their own, or elements placed in the building instead of on a storey. The model summary points these out before you take any sum.
05Where moddl deliberately stops
- moddl doesn’t apply billing rules such as VOB/C. Where a quantity does rest on a rule, for example the outer face of an external wall run to the corners, it’s marked as an assumption and you see in m² how far it deviates.
- If the file has no density, moddl only works out a mass by two named rules and marks it as an assumption: steel with a catalogue section at 7850 kg/m³, concrete with a strength class at 24 kN/m³. Otherwise the mass stays open, because moddl won’t infer a density from a material name.
- Reading and measuring change nothing in the model.
06Try it on your own model
Start with the quantity you need right now anyway, per element, class, material or storey. Open the viewer, load your IFC model and connect your AI, for example Claude or ChatGPT. Then you’ll see straight away where that one figure comes from.
If you then want to change something in the model, your AI tells you beforehand which elements are affected, and moddl reports afterwards what changed. You can take a change back with undo.
Every change creates a new version. If you don’t like a result, restore the version before it in the viewer.
Frequently asked questions
What are the requirements?
You need an AI assistant with MCP support, for example Claude or ChatGPT. ChatGPT can also be used on the free plan. You connect it with your own account, without an API key: Edit BIM with your own AI subscription.
How do I connect Claude or ChatGPT?
moddl walks you through the connection step by step in the viewer. You can also read the guides here: Connect Claude to moddl and Connect ChatGPT to moddl.
What does it cost?
During the beta phase, expected to run until June 30, 2027, moddl is free of charge. We will usually announce the end of the beta at least two weeks in advance. Your AI assistant is subject to your provider's terms.
Does the AI get access to my IFC file?
No. The AI receives only the answers to its requests and reports on the changes it has made. These answers may contain content from the model, such as element names and properties.
Which IFC versions are supported?
IFC4 and IFC4X3. Files in IFC2X3 format are converted automatically on upload.
Can changes be undone?
Yes. Every change creates a new version, and every earlier version can be restored. The uploaded original file stays unchanged.