// AI TOOLS · CARTESIAN BY FORMAS

Cartesian by Formas

Formas' AI 3D-modeling tool that turns words, sketches, and dragged-in references into precise, editable models — native NURBS solids aimed at architecture and product design, not polygon-mesh blobs.

Last verified · 2026-09-15 · by Moe Ameen

What Cartesian by Formas is

Cartesian is an AI 3D-modeling tool from Formas (formas.ai), a design-software company with offices in Singapore and Tokyo. Its pitch is "say it, sketch it, drag it in" — you describe an intent in natural language, sketch a form, or drop in a photo, plan, or scan, and Cartesian builds a 3D model from it. It's aimed at architecture and product design rather than at hobbyist or game-asset 3D, and it's the CAD-adjacent sibling to Formas Studio, the company's separate AI rendering and visualization workspace.

What sets Cartesian apart from the crowded text-to-3D field is precision. Formas positions it to generate native NURBS solids and exact geometry with clean topology and editable assemblies — not the loose polygon meshes most AI 3D generators produce — and to hold the exact geometry of specified elements while leaving the rest flexible. It's built with BIM-style workflows in mind (named elements with explicit relationships), and it exports to SketchUp (.SKP), Rhino (.3DM), and STL. AutoCAD and IFC support are listed as planned.

Cartesian's preview opens September 18, 2026, with early access going first to Formas' founding-circle and studio members and a public waitlist for everyone else. Formas hasn't published pricing for Cartesian, and its geometry-quality claims are vendor-stated rather than independently benchmarked, so treat the specifics as things to confirm once the preview opens. The use cases Formas lists are broad within design: product, furniture, interiors, residential and urban architecture, landscape, manufacturing, 3D printing, and game assets.

Honest framing: Cartesian is a modeling tool. It produces 3D files you can measure, cut, print, or machine, and hand to the rest of a CAD pipeline — but it writes no copy, cuts no video, designs no social graphics, and publishes nothing. Turning a design into content that reaches an audience is a separate job.

What you can make with it

  • Precise, editable 3D models from a text description, a rough sketch, or a dragged-in photo, plan, or scan
  • Native NURBS solids with clean topology for product, furniture, and architectural work
  • Models you can export to Rhino (.3DM), SketchUp (.SKP), and STL for the rest of your pipeline
  • STL-ready geometry for 3D printing and manufacturing
  • BIM-minded assemblies with named elements and explicit relationships
  • Screen-captured turntables and orbit views of a model — the raw visual material for a reveal

How Kompozy turns Cartesian by Formas output into content

Think of Cartesian as the maker's bench and [Kompozy](/) as the launch team. If you sell 3D-printed products, run a furniture line, or ship a product design, Cartesian gets you to a clean, precise model — but a model in a .3DM file doesn't sell anything until people see it. Kompozy is a full content generation and publishing engine, so it takes the visual output of that model and turns a single design into a product-launch content set that actually goes out the door.

The concrete handoff: screen-record a turntable of your Cartesian model, or export a render, and drop it into Kompozy's Quick Ingest. It becomes a vertical Listicle Video or [Clipped Shorts](/glossary/clipped-short) reveal with burned-in branded captions, reframed for Reels, Shorts, and TikTok. From the same source Kompozy fans out the rest of a launch: an Infographic Photo laying out the specs and dimensions, Photo Posts staging the piece, a build-in-public process Carousel via [HyperFrames](/glossary/hyperframes), Quote Graphics, native Text Posts, a product-story Blog Article, and an Email Newsletter to your buyers — all held to one voice by a [Persona Brief](/glossary/persona-brief), then scheduled and published across the eight social platforms plus blog and email through [Autopilot](/glossary/autopilot), behind a per-post review gate. One model in Cartesian becomes a week of on-brand launch content in Kompozy — the formats a modeler was never built to make.

  1. Model your product or design in Cartesian from a prompt, sketch, or reference, and refine the geometry.
  2. Capture the visual: screen-record a turntable or orbit of the model, or export a still/render.
  3. Drop that clip or image into Kompozy Quick Ingest as the source for a new content unit.
  4. Set your Persona Brief once, then generate the launch set — a listicle/clipped reveal, a spec infographic, a build-in-public carousel, photo posts, a product-story blog, and a newsletter.
  5. Review each piece behind the per-post gate, then schedule and publish across the eight social platforms plus blog and email with Autopilot.

Frequently asked questions

What is Cartesian by Formas?

Cartesian is an AI 3D-modeling tool from Formas (formas.ai) that turns natural-language descriptions, sketches, and dragged-in references into editable 3D models. It targets architecture and product design, claims native NURBS solids with clean topology, and exports to SketchUp (.SKP), Rhino (.3DM), and STL. AutoCAD and IFC are listed as planned.

Is Cartesian available yet, and what does it cost?

Cartesian's preview opens September 18, 2026, with early access going first to Formas' founding-circle and studio members and a public waitlist for others. Formas hasn't published pricing for Cartesian at this time, so check the Formas site for current availability and terms.

How is Cartesian different from mesh-based AI 3D generators like Meshy or Tripo?

Most text-to-3D tools output polygon meshes aimed at game assets or quick visuals. Cartesian claims native NURBS solids and precise, editable geometry for CAD and BIM pipelines — solids you can measure, cut, and machine rather than meshes you'd rebuild. That precision claim is vendor-stated in preview, so verify it on your own output.

Can Cartesian create social posts or marketing video?

No. Cartesian builds 3D models and geometry — it makes no captions, social video, carousels, blogs, or scheduled posts. To turn a model or render into published marketing across platforms, you use a content engine like Kompozy, which clips, captions, builds every format, and publishes across the eight social platforms plus blog and email.

Related tools

  • Google MixboardA prompt-driven, Pinterest-style AI concepting board from Google Labs, powered by the Nano Banana image model — being retired on September 28, 2026.
  • MidjourneyThe text-to-image generator known for aesthetic quality and art direction — now also building a separate medical-imaging division.
  • Krea 2Krea AI's first in-house foundation image model, built for aesthetic range and precise style control.
  • Flux 3Black Forest Labs' multimodal foundation model — one system trained jointly on image, video, and audio, plus a robotics action head.
  • Seedream 5.0 ProByteDance's multimodal image model that reasons over a brief, renders dense text, and separates a finished image into editable layers.

← All AI tools · Get started →