What IFC versions are supported?

IFC2X3, IFC4, IFC4X3 — anything ifcopenshell can read. Older or vendor-extended schemas may have unparsed properties; report them and we'll widen coverage.

How big can the file be?

500 MB on cloud, 2 GB on the desktop SKU. The parser runs as a subprocess so a corrupt or oversized file can't bring down the backend.

Is the geometry rendered in 3D?

Phase 1 is queryable only — no embedded 3D viewer for direct uploads yet. Speckle and Autodesk APS sources both render in 3D today; a glTF-based viewer for direct IFC uploads is planned.

Can I update the source with a new file?

Yes — re-upload from the source detail page. The source.id stays the same, the SQLite is overwritten with fresh data. Useful when your architect ships v2.

Does this work offline / air-gapped?

On the iDBQuery Desktop SKU, yes. The parse runs locally; no cloud round-trip. Files never leave the laptop.

What about IFC write-back?

Out of scope — IFC files are treated as read-only ground truth. Edits stay in your authoring tool (Revit, Archicad, Tekla, etc.).

BIM

IFC file (direct upload)

Drag-drop a .ifc — query the model in plain English. No account, no Speckle, no Autodesk.

Connect free Talk to us

Overview

Why teams connect IFC file (direct upload) to iDBQuery

iDBQuery's direct IFC integration lets you skip the BIM-platform tax. Drag a .ifc file (up to 500 MB) into the connect form, and within a minute every IfcWall, IfcDoor, IfcSlab, IfcSpace and IfcBuildingStorey is queryable in plain English. Properties live in the same speckle_objects schema we use for Speckle, so every Speckle chat capability lights up automatically — list ifc_types, count by floor, sum areas, dig into Pset_WallCommon via json_extract. Files stay in your iDBQuery workspace; nothing leaves to a third-party service.

Capabilities

What you can do

Everything iDBQuery's IFC file (direct upload) connector supports out of the box.

Drag-drop .ifc upload, up to 500 MB per file
Parsed by ifcopenshell (IFC2X3 / IFC4 / IFC4X3)
Element-level sync into a per-source SQLite (same schema as Speckle)
Hot-path columns: ifc_type, level, area, volume, length, height, units, application_id (IFC GlobalId)
Property Sets queryable via json_extract on properties_json
Storey/level resolution via IfcRelContainedInSpatialStructure
Subprocess parser — file errors can't crash the backend
Background-thread sync with live progress bar
Re-upload to refresh — same source.id, fresh data
No external account required — fully self-hosted analytic surface

Setup

Get connected in minutes

Pick a project

Project → Add source → IFC file. Or upload directly into your default project from the dashboard.

Drop the file

Drag-drop your .ifc or click to pick. Files up to 500 MB are accepted.

Wait for parse

iDBQuery walks every IfcElement and writes hot-path columns. Typical 50 MB IFC parses in 30-60 s; 200 MB in ~2 min.

Chat the model

Ask 'how many doors?', 'wall area on level L2?', 'list IFC types', 'show me load-bearing walls'. Same chat experience as Speckle sources.

Try this

Sample prompts

>How many windows are in this model?
>Which level has the most slabs?
>Total wall area on Level 02
>List every IFC type with element count
>Find all walls thicker than 200 mm
>What Property Sets are attached to doors?
>How many storeys does this building have?

Frequently asked questions

Speckle is great for live coordination across a team; the IFC route is for one-off audits, models received from a vendor, or air-gapped use where you don't want geometry leaving your iDBQuery workspace. Both can co-exist on the same project.

Keep exploring

Chat with data Reports All integrations Construction use case

Connect IFC file (direct upload) in under a minute.

Free tier covers most teams. No credit card.

Start free Contact us

No credit card required · 1M tokens / month free

Loading…

Frequently asked questions

IIFC file (direct upload)

Why teams connect IFC file (direct upload) to iDBQuery

What you can do

Get connected in minutes

Pick a project

Drop the file

Wait for parse

Chat the model

Sample prompts

Frequently asked questions

Keep exploring

Connect IFC file (direct upload) in under a minute.

IIFC file (direct upload)

Why teams connect IFC file (direct upload) to iDBQuery

What you can do

Get connected in minutes

Pick a project

Drop the file

Wait for parse

Chat the model

Sample prompts

Frequently asked questions

Keep exploring

Connect IFC file (direct upload) in under a minute.

IFC file (direct upload)

IFC file (direct upload)