NonicaTab actually started back in 2020 as a Revit Add-In to make custom Revit toolbars easier, we ended up building over 36 tools that people can mix and match depending on what they need. Then, in early 2025, when MCPs [Model Context Protocols] came out, we saw an opportunity and shifted toward AI, specifically, an A.I. Connector for Revit that lets AI extract, review, report on, and document your models using natural language.
We offer it in two versions: there is a Free version, which includes a solid set of our most-used tools plus a limited version of the AI Connector, so people can really get a feel for what it does. And then there is PRO, which unlocks the full toolset and unlimited use of the AI Connector for those who end up relying on it more heavily.
 |
| Jaime Alonso |
On the toolbar side, Free includes a curated set of our most popular tools, the ones people reach for most often [13 tools]. PRO gives you all 36+ tools, so if there's something more specific or niche you need, it's probably in there.
On the AI Connector side, Free lets you use it too, with some usage limits, so you can genuinely try it out on real work and see how it fits into your workflow. It is mostly geared toward extracting, reviewing, and reporting from Revit, asking questions, pulling data, generating reports. The editing side, like setting parameters or adding views to sheets, is reserved for PRO.
So honestly, a lot of users do just fine on Free, especially if they are mostly doing lighter documentation and review work. PRO makes more sense once you want AI to actually make changes in the model, not just report on it.
What popular 3rd party add-ins are baked into the Pro version?
A few tools were inspired from some well-known third-party add-ins such as Align or Room Finishing. For instance, for the tool Color Elements by Parameters, we converted the Color Splasher from C# to Python, and added a few features (e.g. the option to create view filters for the colors or automatic legends). After releasing in Nonicatab, it was also included in pyRevit set of tools, so especially proud about that one. The idea was to build native versions inside NonicaTab that we could keep supporting long-term, so users aren't stuck depending on something that might not get another update.
Can you please describe in detail your AI integration?
The A.I. Connector is a secure bridge (MCP) between Revit and AI desktop apps like Claude or ChatGPT Work, giving A.I. over 50 pre-built, tested tools for exploring, analyzing, editing, and documenting Revit models.
The key idea behind it is that traditional A.I. chats for Revit generated and run code from scratch, which tends to lead to errors and breaks down on complex tasks. So instead of letting the AI write its own Revit API code on the fly, we give it a toolbox of tools we have already built and tested, things like reading parameter values, checking family file sizes, seeing which views are placed on which sheets, and so on. The A.I. picks the right tool for the job instead of improvising code, which makes the results more reliable and stable.
In practice, that means someone can just ask, something like "review all doors in this project for accessibility compliance" or "generate a report on model warnings," and the AI uses those tools directly inside the live Revit session to pull real data and give a real answer, no scripting, no menus.
On setup, it's meant to be simple, no dependencies like Node.js, everything installs automatically through the A.I. Connector.
Can you describe a few useful implementations of your AI on a Revit project?
A few examples; Reviewing and fixing parameter formatting across a model, cleaning up view templates and re-assign scope boxes, or renaming and renumbering views, rooms, and sheets to match a standard. It can also translate parameters or element names, generate interactive reports with charts connected to Revit, and break down warnings by severity and which elements they affect.
Beyond review, it is also useful for action items, selecting or coloring elements by category or parameter, creating sheets, sections, levels, or grids, assigning revisions in bulk, tagging elements by type or condition, printing to PDF with custom naming, or isolating elements tied to a specific warning or parameter.
If I had to sum it up, most of what people actually use it for day-to-day is QA/QC; catching inconsistencies or formatting issues, missing/duplicated tags, wrong or empty parameters, or warnings before they become real problems, instead of relying on someone manually clicking through the model.
What is the price of the Pro version?
The price is 85€ per user per year, roughly 7€ per month. For initial testing, I would still recommend to start with the Free version or the free trial of the PRO version.
Is there anything else you'd like our readers to know about your add-ins?
Just that we are constantly trying to improve this, and a lot of it comes from things we learn once teams start using it in projects.
One example: we noticed that when the AI had to work with big lists of elements, it was slow and it was consuming a lot of tokens just to handle the ids. So we introduced the subsets, a way to group and reference big sets of elements as one single reference, instead of passing every id back and forward, and this alone made a lot of the workflows 10x more efficient.
The same happened with the reports. In the beginning, the reports were static, AI generates it and that's all. But we hear from the users that they want to interact with the data after, so now the reports stay live-connected with the model, and you can click or edit Revit elements directly from the report, without the need to spend more tokens asking AI every time.
These kind of things come directly from the feedback and from watching where is the friction for the people that are really using it, not from a roadmap that we make by ourselves.
Thanks Jaime!
Advertisement
No comments:
Post a Comment