Automate IFC processing online — batch pipelines, no scripting
BIMCamel's Automate tab turns any of the 40+ browser tools into steps in a repeatable pipeline. Drag a Clean → Validate → Merge → Convert chain together once, run it on a single IFC or a whole folder of models, and re-run it on every future revision without clicking through the steps again. It is IFC batch processing online — the BIM automation workflow you used to script with Python and desktop APIs, rebuilt as drag-and-drop in the browser.
There is nothing to install and nothing to license per seat. Guests can chain up to 3 steps and run 3 pipelines a day without an account; a free account raises that to 5 steps and 10 runs a day, saves your pipelines for one-click re-runs, and keeps a run history.
Build a pipeline from 40+ operations
Every tool in the BIMCamel catalogue is available as a pipeline step: cleanup (remove elements, strip property sets, deep purge), optimisation (compress, bounding-box simplification), organisation (merge, split, storey reassignment, schema upgrade), validation (health report, clash detection, IDS check and the rest of the checks catalog), data extraction (IFC to Excel, custom reports, COBie), and format conversion (glTF, OBJ, STEP, CAD drawings). Steps are reordered by drag-and-drop, and each step opens its own configuration with sensible defaults.
On top of the tools themselves, the builder adds pipeline-only control steps that never appear in the normal tool grid: conditional gates, abort-on-failure guards, element-count assertions, bulk renaming, log messages, and filter-to-selection.
Start from a template
The templates gallery seeds the builder with a curated chain in one click — open one, tweak the steps, run it:
- The Ultra-Compressor — strip non-geometry, simplify to bounding boxes, purge orphans.
- The Privacy Sanitizer — remove property sets, materials, and external references before sharing.
- The Model Health Fixer — run the repair suite, then upgrade the schema.
- IDS Gate — validate against an IDS ruleset and stop the run if the model fails.
- Repair only if checks fail — a conditional chain that skips the repair step on clean models.
- Coordination pack — clash detection plus a full health report in one run.
Quick tasks vs pipelines
A quick task is a single tool run — upload, configure, download. A pipeline chains several operations into one run, passing each step's output to the next, so a five-tool cleanup becomes one upload and one download. Both run on the same processing engine and both count as one run against your daily allowance per execution, so the pipeline route is also the cheaper way to spend your quota when a job needs several tools.
Save pipelines and re-run them
Any signed-in user — including the free tier — can name and save a pipeline. Saved pipelines live in your account on the Automate tab next to your run history: re-run one on a new upload in two clicks, or use "Re-run from here" to replay a past run from a specific step onward instead of repeating the whole chain. Run history keeps your 10 most recent runs on the free tier and 500 on Pro and Team, each with per-step logs, status, and output downloads.
Quality gates: what happens when a step fails
Validation steps (IDS check, clash detection, assertions) record their findings but do not stop the run by themselves — a model that fails an IDS check still flows to the following steps unless you tell the pipeline otherwise. Two control steps make the failure behaviour explicit:
- Abort on failure — hard-fails the run when the previous validation step reported failures. By default any failure aborts; set a percentage threshold to tolerate small failure rates and abort only above it.
- Conditional gate — branches on the last validation result (previous step passed / failed, none failed so far / any failed so far). When the condition isn't met it can stop the run gracefully (the model processed so far is still saved), skip the next step, or skip the next N steps.
- Assert element count — checks that a class count is inside an expected range; in strict mode an out-of-range count fails the run, in soft mode it's recorded as a finding and the run continues.
Honest limits, per tier
Every tool is available on every tier — the caps are on quantity, not capability:
- Guest (no account): 3 runs a day, 25 MB uploads, up to 3 steps per pipeline, stream-only outputs (deleted after download), no run history.
- Free account: 10 runs a day, 100 MB uploads, up to 5 steps per pipeline, saved pipelines, 7-day retention, last 10 runs in history.
- Pro (€19/mo): 200 runs a day, 500 MB uploads, unlimited pipeline length, unlimited retention, email-when-ready notifications, 500-run history.
- Team (€49/mo): 400 pooled runs a day across 5 seats, 1 GB uploads, and email notifications to teammates and clients — up to 15 recipients per job.
Frequently asked questions
Which operations can be chained in a pipeline?
Nearly the whole catalogue — 40+ operations spanning cleanup, optimisation, merge/split, validation (health, clash, IDS and the other checks), property editing, data extraction, and format conversion — plus pipeline-only control steps like conditional gates and abort-on-failure guards. A handful of interactive tools (the IDS Maker, the two-model compare view, the viewer/inspector) run as standalone pages instead of pipeline steps.
Does a failed check stop the whole run?
Not by default. Validation steps record their findings and the run continues — so a report still comes out even when the model has problems. To enforce a gate, add an Abort-on-failure step (hard-fails the run when the previous validation reported failures, with an optional percentage threshold) or a Conditional gate (stops gracefully or skips the next step(s) based on whether previous checks passed).
Can I save a pipeline and re-run it later?
Yes. Any signed-in user — including the free tier — can save named pipelines and re-run them on future uploads. The Automate tab also keeps a run history with per-step logs, and a "Re-run from here" action that replays a past run from a chosen step onward.
What are the file size and daily run limits?
Guests get 3 runs a day with 25 MB uploads and pipelines up to 3 steps. A free account raises that to 10 runs a day, 100 MB uploads, and 5 steps. Pro allows 200 runs a day with 500 MB uploads and unlimited pipeline length; Team pools 400 runs a day across 5 seats with 1 GB uploads.
Can I batch-process a whole folder of IFC files?
Yes — drop a folder onto a pipeline run and every IFC queues with its own row, progress, and output link. Outputs land in your workspace one entry per file, so each per-file run stays replayable later.
Can pipelines run on a schedule or from a webhook?
Not today. Pipeline runs start from your browser — but because saved pipelines re-run on any new upload in two clicks, the recurring-workflow case is covered without scripting: upload the new revision, pick the saved pipeline, run.
Do I need an account to try it?
No. Guests can open the builder, chain up to 3 steps, and run 3 pipelines a day on files up to 25 MB. You'll need a free account to save pipelines and keep run history, since both need somewhere to live.
What happens to my file after processing?
Your file is uploaded over HTTPS and processed on our servers. Guest uploads are stream-only — the working files are deleted as soon as your download finishes. Free accounts keep results for 7 days; Pro keeps them 90 days and Team 180 days. Files are never shared with third parties or used to train AI.